the determination of cis and trans fatty acid isomers in partially
Post on 12-Feb-2022
2 Views
Preview:
TRANSCRIPT
The Determination of Cis and Trans Fatty Acid Isomers in Partially Hydrogenated Plant Oils
By
Christiaan De Wet Marais
Thesis presented in partial fulfillment of a Masters degree in Chemistry (Analytical)
at the
Department of Chemistry and Polymer Science
University of Stellenbosch
Study leader: Prof. A.M. Crouch Department of Chemistry and Polymer Science University of Stellenbosch Co- study leader: Dr. C.M. Smuts Nutritional Intervention Research Unit The Medical Research Council Tygerberg
March 2007
DECLARATION
I, Christiaan DeWet Marais, hereby declare that the work
contained in this thesis is my own original work and that I have
not previously in its entirety or in part submitted it at any
university for a degree.
Signature: Date:
ii
ABSTRACT Trans isomers are formed during the partial hydrogenation process of cis unsaturated fatty
acids. The major source of trans fatty acids in the normal person's diet is from margarines and
shortenings made from these partially hydrogenated plant and marine oils. In addition to
influencing lipid risk factors for cardiovascular disease, trans fatty acids have also been
implicated in breast cancer, and in poor fetal development and reduced early infant growth. In
reality, trans fatty acids have been consumed for centuries, since they occur naturally in beef,
mutton, butter, milk and other dairy products. Though it has been shown that these naturally
occurring trans fatty acids have different effects on the health of humans. With the
implementation of the new labelling law in South Africa, the trans fatty acids content of food
items must be displayed on the food label. Therefore, it becomes necessary to optimise the
analytical methodology for the determination of trans fatty acids in foods.
Many publications have reported on the quantification of the total concentration of trans fatty
acids in food samples, while less work has been done on the identification and quantification
of the different cis and trans unsaturated fatty acid isomers found in foods made from
partially hydrogenated oils. The objective of this study was to standardise and optimise an
analytical technique to identify and quantify the different cis and trans mono-unsaturated fatty
acid isomers in local margarines and bread spreads.
Seeing that fatty acids are the group of lipids most commonly analysed by GLC and the
availability of highly polar capillary columns bonded with cyanoalkyl polysiloxan phases, it
was decided to use GLC for the identification and quantification of the different cis and trans
isomers in a selected group of margarines. It was further decided to evaluate two BPX-70
capillary columns packed with cyanoalkyl polysiloxan phases. The one a 30 m BPX-70
capillary column, normally used for routine fatty acid analyses, and the other a 120 m BPX-70
capillary column.
To extract the fatty acids from the samples, extraction solutions including chloroform,
methanol and hexane were evaluated. For the transmethylation of the extracted fatty acids 0.5
M sodium methoxide in methanol and 5% concentrated sulphuric acid in methanol, were
evaluated.
iii
To optimise the GLC conditions, different column temperature programs and column gas flow
rates were applied.
Of the three different extraction solutions evaluated in this study, chloroform/methanol (2:1)
solution gave the best fatty acid recovery. It was also found that the 5% concentrated
sulphuric acid/methanol transmethylation solution, gave a 7% better FAME recovery than 0.5
M sodium methoxide/methanol. When analysing a pooled margarine sample, it was found that
with a 30 m BPX-70 capillary column the different cis and trans 18:1 isomers were forced to
overlap due to the narrow elution gap, while a 120 m BPX-70 column provided the required
mechanism for extending the retention times of the different isomers by retaining the different
compounds longer. In this way, the retention times of the different isomers were pulled apart,
and a greater separation space was available to identify more different isomers. It was found
that column temperature had a major effect on the separation power of the 120 m BPX-70
capillary column. Isothermal operation at 181oC produced the fewest overlapping peaks and 5
peaks could be separated before the main cis-9, 18:1 isomer and 7 peaks thereafter. Isothermal
temperatures above and below 181oC produce some additional overlapping problems.
The use of Ag-TLC separation before GLC analyses improves the identification of the
different isomers, but it could not separate all the isomers, with the same geometrical structure
that are eluting close together.
Using the optimised GLC conditions, eighteen different margarines were analysed. The
results show that the normal occurring fatty acids, as well as most of the cis and trans fatty
acids can be identified and quantified in one analytical run. The results further show that the
trans fatty acid content of the selective group of local margarines are not as high as reported
for some other countries, but that the saturated fatty acid content of these margarines is higher
than the recommended levels.
Capillary electrophoresis was also utilised, but the separation and identification of the cis and
trans fatty acid isomers in a standard sample were unsuccessful and much more analytical
development is needed.
iv
OPSOMMING Trans isomere word gevorm tydens die gedeeltelike hidrogenering van cis onversadigde
vetsure. Die hoofbron van trans vetsure in die normale persoon se dieet word gevind in
margarine en bakvet wat van gedeeltelik gehidrogeneerde plant en mariene olies vervaardig
word. Buiten die effek wat trans vetsure op die lipied risiko faktore vir kardiovaskulêre
siektes het, word dit verder verbind met borskanker, swak fetale ontwikkeling en vertraagde
groei in die jong kind. In werklikheid word trans vetsure reeds vir eeue ingeneem aangesien
hulle natuurlik in bees- en skaapvleis, botter, melk en ander suiwelprodukte voorkom. Daar is
egter getoon dat hierdie natuurlike trans vetsure verskillende uitwerkings op die mens se
gesondheid het. Die nuwe Wet op Etiketering in Suid-Afrika vereis dat die trans vetsuur
inhoud van voedselitems op die voedseletiket vertoon moet word. Dit het daarom nodig
geword om die analitiese metodologie vir die bepaling van trans vetsure in voedsels te
optimaliseer.
Baie publikasies het al gerapporteer oor die bepaling van die totale konsentrasie van die trans
vetsure in voedsel monsters, maar minder werk was gedoen op die identifisering en
kwantifisering van die verskillende cis en trans onversadigde vetsuur isomere in voedsels wat
vervaardig word van gedeeltelike hidrogeneerde plantolies. Die doel van hierdie studie was
om ‘n analitiese tegniek te standardiseer en optimaliseer vir die identifisering en
kwantifisering van die verskillende cis en trans mono-onversadigde vetsuur isomere in
margarine en smere.
Omdat vetsure die groep lipiede is wat die mees algemeen deur GLC geanaliseer word, en
omdat lang, hoogs polêre kapillêre kolomme, gebind met siano-alkiel polisiloksaan fases
geredelik beskikbaar is, was daar besluit om GLC te gebruik vir die identifisering en
kwantifisering van die verskillende cis en trans isomere in ’n uitgesoekte groep margarines.
Daar was ook besluit om twee BPX-70 kapillêre kolomme, wat gepak is met siano-alkiel
polisiloksaan fases, te evalueer. Die een, ’n 30 m BPX-70 kapillêre kolom, wat normaalweg
vir roetine vetsuurbepalings gebruik word, en die ander, ’n 120 m BPX-70 kapillêre kolom.
Vir die vetsure ekstraheering van die monsters, ekstraksie oplossings wat insluit chloroform,
metanol en hexaan was geevalueer. Vir die transmetelering van die geekstraheerde vetsure,
v
0.5 M natrium methoxide in metanol en 5% gekonsentreerde swaelsuur in methanol was
geevalueer. Om die GLC kondisies te optimaliseer, verskillende kolom temperature en kolom
gas vloei spoed, was getoets vir die analiseering van die margarine monsters.
Van die drie ekstraksie metodes wat geevalueer was het ‘n oplossing van chloroform/metanol
(2:1) die beste vetsuur herwinning gegee. Daar is ook gevind dat 5% gekonsentreerde
swaelsuur in metanol ‘n 7% beter herwinning van vetsuur metiel esters gegee het as 0.5 M
natrium methoxide in methanol.
In ’n saamgestelde margarine monster is gevind dat met ’n 30 m kolom die verskillende cis en
trans 18:1 isomere geforseer word om te oorvleuel as gevolg van die nou elueringsgaping,
terwyl ’n 120 m kolom die kapasiteit het om die retensietye van die verskillende isomere te
verleng deur die verskillende komponente langer terug te hou. Op hierdie manier is die
retensietye van die verskillende isomere uitmekaar getrek, en is ‘n groter skeidingspasie
beskikbaar vir die verskillende isomere. Hierdie skeidingskrag bring mee dat meer isomere
geïdentifiseer kan word. Daar was gevind dat kolom temperatuur ‘n groot effek op die
skeidingsvemoë van ‘n 120 m kapillêre kolom het. Met ‘n isotermiese temperatuur van 181oC
het die minste pieke geoorvleul en kon 5 pieke voor die hoof cis-9, 18:1 isomeer geskei word
en 7 pieke daarna. Isotermiese temperature hoër en laer as 181oC het additionele
oorvleulingsprobleme veroorsaak.
Die gebruik van Ag dun-laagchromatografiese skeiding voor GLC analise, verbeter die
identifisering van die verskillende isomere, maar kon nie al die isomere met dieselfde
geometriese struktuur wat na aanmekaar elueer skei nie. Hierdie is as gevolg van die klein
verskil in hulle onderskeie retensietye.
Deur gebruik te maak van die geoptimaliseerde GLC kondisies, is agtien verskillende
margarines ontleed. Die resultate toon dat die vetsure wat gewoonlik voorkom, asook meeste
van die cis en trans vetsure in een analitiese sessie geïdentifiseer en gekwantifiseer kan word.
Die resultate toon verder dat die trans vetsuur inhoud van die geselekteerde groep plaaslike
margarines nie so hoog is soos gerapporteer vir sommige ander lande nie, maar dat die
versadige vetsuurinhoud van hierdie margarines hoër is as die aanbevole vlakke.
vi
Kapillêre elektroforese is ook gebruik, maar die skeiding en identifisering van die cis en trans
vetsuur-isomere in ’n standaardmonster was nie suksesvol nie, en verdere analitiese
ontwikkeling word benodig.
vii
CONTENTS Page Abstract iii
Opsomming v
Acknowledgements xi
List of Figures xii
List of Tables xvi
List of Graphs xviii
List of Abbreviations xix
CHAPTER 1 INTRODUCTION AND AIMS OF THE STUDY 1 1.1 Introduction 1
1.2 Aims of the study 5
1.2.1 Specific objectives 6
CHAPTER 2 LITERATURE REVIEW 7 2.1 Introduction 7
2.2 Fatty acids 7
2.3 Trans fatty acids 11
2.3.1 Natural occurring trans fatty acids 11
2.3.2 Commercially produced trans fatty acids 12
2.4 Hydrogenation 14
2.5 Naming of fatty acids 16
2.6 Analytical procedures for the determination of cis and trans fatty acids 20
2.6.1 Introduction 20
2.6.2 Infrared spectroscopy 20
2.6.3 Silver impregnated thin layer chromatography 21
2.6.4 High performance liquid chromatography 21
2.6.5 Gas liquid chromatography 22
2.6.6 Capillary electrophoresis 23
2.7 Lipid extraction 23
viii
2.8 Transmethylation of fatty acid 25
2.8.1 Acid-catalysed transmethylation 26
2.8.2 Base-catalysed transmethylation 27
2.9 Instrumentation 28
2.9.1 Gas liquid chromatograph 28
2.9.1.1 Introduction 28
2.9.1.2 Gas liquid chromatograph instrument 29
2.9.1.3 Carrier gases 30
2.9.1.4 Columns 31
2.9.1.5 Column temperature 31
2.9.1.6 Injectors 32
2.9.1.7 Detectors 33
2.9.2 Capillary electrophoresis 34
2.9.2.1 Introduction 34
2.9.2.2 Capillary zone electrophoresis 37
2.9.2.3 Micellar electrokinetic chromatography 37
2.9.2.4 Capillary electrophoresis instrument 38
2.9.2.5 Capillaries 39
2.9.2.6 Electrolyte system 39
2.9.2.7 Sample introduction 40
2.9.2.8 Detectors 40
CHAPTER 3 MATERIALS AND METHODS FOR GLC 41 3.1 Introduction 41
3.2 Sampling and sample handling 41
3.3 Chemicals and gases 41
3.4 Evaluation of different lipid extraction solutions 42
3.5 Evaluation of lipid transmethylation procedures 43
3.6 Sample preparation 44
3.7 Evaluation of the two BPX-70 GLC columns 45
3.8 Identification of the different standard isomers 46
3.9 Evaluation of silver ion thin layer chromatography 46
ix
CHAPTER 4 GLC RESULTS AND DISCUSSION 48 4.1 Evaluation of the different extraction solvents 48
4.2 Evaluation of the two transmethylation solvents 56
4.3 Evaluation of the two columns 64
4.4 Identification of the standard isomers 66
4.5 Evaluation of different column temperatures 70
4.6 Results of silver ion thin layer chromatography 90
4.7 Results of the margarine samples 95
CHAPTER 5 MATERIALS AND METHODS FOR CE 101 5.1 Introduction 101
5.2 Samples 101
5.3 Reagents and solutions 102
5.4 Instrumentation 102
CHAPTER 6 CE RESULTS AND DISCUSSION 103 CHAPTER 7 CONCLUSIONS 105 CHAPTER 8 REFERENCES 108
x
ACKNOWLEDGEMENTS
I sincerely wish to express my gratitude to the following people:
• Dr A. Dhansay, Director of the Nutritional Intervention Research Unit of the Medical
Research Council: for his encouragement and support to do this M.Sc. course.
• Prof A. Crouch, of the Department of Chemistry and Polymer Science, University of
Stellenbosch: my study leader for his guidance and advice with the preparation of this
thesis.
• Dr M. Smuts, of the Nutritional Intervention Research Unit of the Medical Research
council: my co-study leader for his guidance, advice and constructive criticism during
the execution of this study and the preparation of this thesis.
• The Medical Research Council: for financial support and the provision of an ideal
research environment to do this study.
• My colleagues at NIRU: for their understanding and support.
• Ms Jean Fourie: for the language editing of this thesis.
• Dr J. Seier and his wife, Sally: for their support and help with editing.
• My wife, Martelle: for her loving support, encouragement and understanding.
• Almighty Father: for the health, motivation and guidance.
xi
LIST OF FIGURES
Page Figure 1 The geometrical structures of trans unsaturated, cis unsaturated and
saturated fatty acids. 3
Figure 2 Geometrical structures of fatty acids with different degrees of saturation. 9
Figure 3 Non-conjugated polyunsaturated fatty acid structure. 9
Figure 4 Conjugated polyunsaturated fatty acid structure. 10
Figure 5 The geometrical structures of the cis and trans mono-unsaturated fatty
acids 13
Figure 6 Geometrical structure of cis-9, cis-12, octadecadienoic acid, also known
as 18:2 (n-6). 18
Figure 7 The chemical structure of a triacylglycerol molecule. 25
Figure 8 Acid-catalysed transmethylation reaction of a lipid to form a methyl ester. 26
Figure 9 Base-catalysed transmethylation reaction of a lipid to form a methyl ester. 27
Figure 10 Basic components of a GLC. 29
Figure 11 Van Deemter plot indicating the effect of the velocity of nitrogen,
helium and hydrogen as a carrier gas on the theoretical plate height. 30
Figure 12 Split/ Splitless injector. 33
Figure 13 Schematic drawing of a flame ionisation detector. 34
Figure 14 Stern’s model of the double-layer charge distribution at a negatively
charged capillary wall leading to the generation of a Zeta potential and
EOF. 35
Figure 15 Flow profiles of EOF and laminar flow. 36
Figure 16 Schematic representation of the arrangement of the main components of
a capillary electrophoresis instrument. 38
Figure 17.1 Chromatogram of test sample and external standard using chloroform:
methanol (2:1) as the extraction solution and a 30 m BPX-70 column. 48
xii
Figure 17.2 Chromatogram of the test sample and external standard using
chloroform: methanol (1:1) as the extraction solution and a 30 m BPX-
70 column. 49
Figure 17.3 Chromatogram of the test sample and external standard using hexane as
the extraction solution and a 30 m BPX-70 column. 49
Figure 17.4 Chromatogram of the test sample and external standard using
chloroform: methanol (2:1) as the extraction solution and a 120 m BPX-
70 column. 50
Figure 17.5 Chromatogram of the test sample and external standard using
chloroform: methanol (1:1) as the extraction solution and a 120 m BPX-
70 column. 51
Figure 17.6 Chromatogram of the test sample and external standard using hexane as
the extraction solution and a 120 m BPX-70 column. 51
Figure 17.7 The chromatogram of the final pooled hexane extractions to verify that
washing the samples three times with hexane recovered all FAME in the
nine test samples. 55
Figure 18.1 Chromatogram of the test sample using 5% concentrated sulphuric acid
in methanol as the transmethylation reagent with a 30 m BPX-70
column 56
Figure 18.2 Chromatogram of the test sample using 0.5 M methoxide/methanol as
the transmethylation reagent, with a 30 m BPX-70 column. 57
Figure 18.3 Chromatogram of the test sample using 5% concentrated sulphuric acid
in methanol as the transmethylation reagent with a 120 m BPX-70
column. 57
Figure 18.4 Chromatogram of the test sample using 0.5 M methoxide/methanol as
the transmethylation reagent with a 120 m BPX-70 column. 58
Figure 19.1 Part of a chromatogram showing the different 18:1 fatty acid isomers
using a 30 m BPX-70 column. 64
Figure 19.2 Part of a chromatogram showing the different 18:1 fatty acid isomers
using a 120 m BPX-70 column. 65
xiii
Figure 20.1 Part of the chromatogram showing the separation of six standard 18:1
isomers analysed at a column temperature of 151°C on a 120 m BPX-70
capillary column. 66
Figure 20.2 Part of the chromatogram showing the separation of six standard 18:1
isomers analysed at a column temperature of 171°C on a 120 m BPX-70
capillary column. 67
Figure 20.3 Part of the chromatogram showing the separation of six standard 18:1
isomers analysed at a column temperature of 181°C on a 120 m BPX-70
capillary column. 67
Figure 20.4 Part of the chromatogram showing the separation of six standard 18:1
isomers analysed at a column temperature of 191°C on a 120 m BPX-70
capillary column. 68
Figure 21.1 Chromatogram of sample analysed at column temperature of 151o C. 71
Figure 21.2 Chromatogram of sample analysed at column temperature of 155oC. 72
Figure 21.3 Chromatogram of sample analysed at column temperature of 160oC. 73
Figure 21.4 Chromatogram of sample analysed at column temperature of 165oC. 74
Figure 21.5 Chromatogram of sample analysed at column temperature of 170oC. 75
Figure 21.6 Chromatogram of sample analysed at column temperature of 175oC. 76
Figure 21.7 Chromatogram of sample analysed at column temperature of 177oC. 77
Figure 21.8 Chromatogram of sample analysed at column temperature of 179oC. 78
Figure 21.9 Chromatogram of sample analysed at column temperature of 181oC. 79
Figure 21.10 Chromatogram of sample analysed at column temperature of 183oC. 80
Figure 21.11 Chromatogram of sample analysed at column temperature of 190oC. 81
Figure 21.12 Chromatogram of sample analysed at column temperature of 197oC. 82
Figure 21.13 Chromatogram of the pooled sample analysed at column temperature of
181oC. 85
xiv
Figure 22.1 Photograph of a TLC plate impregnated with 10% (w/v) silver nitrate
showing the separation of the cis and trans mono-unsaturated FAME
isomer fractions in the pooled margarine sample. 90
Figure 22.2 Part of the chromatogram of the trans mono-unsaturated FA fraction of
the pooled sample, after Ag-TLC separation. The fraction was analysed
with a 120 m BPX-70 capillary column at a column temperature of
181°C. 91
Figure 22.3 Part of the chromatogram of the cis mono-unsaturated fraction of the
pooled sample, after Ag-TLC separation. The fraction was analysed
with a 120 m BPX-70 capillary column at a column temperature of
181°C. 93
Figure 23.1 Part of the GLC chromatogram of sample P showing the 18:0 (A) 18:1
(B) and 18:2 (C) fatty acids, as well as a cis-11 isomer. 99
Figure 23.2 Part of the GLC chromatogram of sample D showing the 18:0 (A), 18:1
(B) and 18:2 (C) fatty acids, as well as the cis and trans 18:1 fatty acid
isomers. 100
xv
LIST OF TABLES Page
Table 1. The scientific names, shorthand designation and trivial names of some
of the fatty acids. 19
Table 2. Recovery results of the test sample, as determined by the area counts of
triplicate extractions using chloroform/methanol (2:1),
chloroform/methanol (1:1) and hexane as the extraction solutions,
injected into a 30 m BPX-70 column. 53
Table 3. Recovery results of the test sample, as determined by the area counts of
triplicate extractions using chloroform/methanol (2:1),
chloroform/methanol (1:1) and hexane as the extraction solutions,
injected into a 120 m BPX-70 column. 53
Table 4. Recovery results as determined by the GLC area counts of triplicate
extractions when using 5% sulphuric acid/methanol and 0.5 M sodium
methoxide/methanol transmethylation solvents and a 30 m BPX-70
column. 58
Table 5. Recovery results as determined by the GLC area counts of triplicate
extractions when using 5% sulphuric acid/methanol and 0.5 M sodium
methoxide/methanol transmethylation solvents and a 120 m BPX-70
column. 59
Table 6. Recovery results of different test samples concentrations, using 5%
sulphuric acid/ methanol and 0.5 M sodium methoxide/ methanol
reagents as the two transmethylation reagents. 62
Table 7. The effect of column temperature on the percentage composition of the
different fatty acid isomers analysed with a 120 m BPX-70 capillary
column 69
xvi
Table 8. The percentage composition of the different identifiable peaks and of the
main 18:1 fatty acid isomer in the pooled margarine sample, using
different column temperature between 151°C and 197°C. 83
Table 9. The total fatty acid concentration in mg/100 g of the pooled margarine
sample injected into a 120 m BPX-70 capillary column at different
column temperatures between 151°C and 197°C. 88
Table 10. The total fatty acid concentration in mg/100 mg of the pooled margarine
sample injected five times into a 120 m BPX-70 capillary column at a
column temperature of 181°C. 89
Table 11. The GLC results (in percentage composition) of the different trans 18:1
isomers, with and without preceding Ag-TLC separation. 92
Table 12. The GLC results (in percentage composition) of the different cis 18:1
isomers, with and without preceding Ag-TLC separation. 94
Table 13. The total fatty acid composition (mg/ 100 mg) of the margarine samples
analysed with a 120 m BPX-70 capillary column at a column
temperature of 181°C and a hydrogen gas flow rate of 30 cm sec-1. 97
Table 14. The sum of the saturated, mono-unsaturated and polyunsaturated fatty
acids (mg/ 100 mg) of the margarine samples analysed with a 120 m
BPX-70 capillary column at a column temperature of 181°C and a
hydrogen gas flow rate of 30 cm sec-1. 98
xvii
LIST OF GRAPHS Page
Graph 1. The percentage recovery of the test sample using a 30 m and a 120 m
BPX-70 capillary column and chloroform:methanol (2:1),
chloroform:methanol (1:1) and hexane as the extraction solutions. 54
Graph 2. The percentage recovery of the test sample using 5% sulphuric acid/
methanol and 0.5 M sodium methoxide/methanol transmethylation
reagents after analysis with two GLCs equipped with 30 m and 120 m
BPX-70 columns 60
Graph 3. The percentage recoveries of different samples, using 5% sulphuric
acid/ methanol and 0.5 M sodium methoxide/ methanol
transmethylation reagents. 63
xviii
LIST OF ABBREVIATIONS
Ag-HPLC High performance liquid chromatography with pack columns
impregnated with silver nitrate
Ag-TLC Silver nitrate impregnated thin layer chromatography
AOAC Association of Official Analytical Chemists
AOCS American Oil Chemists’ Society
Avg Average
Avg Rec Average recovery
Boron trifluoride BF3
BHT Butylated hydroxytoluene
c Cis
CHD Coronary heart disease
CE Capillary Electrophoresis
CEC Capillary electrochemistry
CLA Conjugated linoleic acids
C:M Chloroform/methanol
Carbon disulfide CS2
CZE Capillary zone electrophoresis
DHA Docosahexaenoic acid
EOF Electroosmotic flow
EPA Eicosapentaenioc acid
FAME Fatty acid methyl ester
FID Flame ionisation detector
FTIR Fourier-transform infrared spectroscopy
GLC Gas liquid chromatography
HDL High-density lipoproteins
HPLC High performance liquid chromatography
H SO Sulphuric acid 2 4
H Hydrogen 2
ID Internal diameter
IR Infrared spectroscopy
xix
IUPAC International Union of Pure and Applied Chemistry
LDL Low-density lipoproteins
Lp (a) Lipoprotein (a)
M Mole
MEKC Micellar electrokinetic chromatography
mM Millimol
MUFA Mono-unsaturated fatty acids
Na Nano ampere
NaOCH Sodium methoxide 3
N Nitrogen 2
TFA Trans fatty acids
TLC Thin layer chromatography
SDS Sodium dodecyl sulphate
STD Standard deviation
t Trans
UV Ultra violet
WCOT Wall-coated open tubular
% CV Percentage coefficient of variance
xx
CHAPTER 1
INTRODUCTION AND AIMS OF THE STUDY
1.1 Introduction
The new labelling law of South Africa that will come into effect during 2006, states that the
total concentration of the trans fatty acids in foods must be correctly displayed on food labels.
(Draft Regulations Relating to Labelling and Advertising of Foodstuffs 2002, no R 1055). The
major source of trans fatty acids in the human diet comes from margarines and shortenings
made from partially hydrogenated plant and marine oils (Katan et al., 1995). It is well known
that trans isomers are formed during the hydrogenation process of cis unsaturated fatty acids
(Sommerfeld, 1983).
With the discovery that saturated fats have adverse effects on blood lipids, people turned to
plant oils as a safe replacement for the saturated animal fats and butters used in cooking and
table spreads. While few would question the health benefits of using some plant oils, it is the
partially hydrogenated oils that have come under fire. The partial hydrogenation of plant oils
improves the stability of the oil and makes it less likely to be oxidised. This process also
converts the oil into a semi-solid fat. During the partial hydrogenation process, a variety of cis
and trans fatty acid isomers are produced. The resulting fats and oils, in addition to containing
trans fatty acids, have reduced amounts of the essential fatty acids, linoleic acid (18:2 n-6)
and alfa-linolenic acid (18:3 n-3) (Emkin, 1995). These essential polyunsaturated fatty acids,
with other unsaturated fatty acids are transformed, because they tend to oxidise easily, causing
the oil to become rancid quite quickly. It has further been postulated that trans fatty acids in
the circulating blood stream also inhibit the conversion of essential fatty acids, thereby
increasing the requirements for essential fatty acid intake (Zevenbergen et al., 1988).
The current pressure to reduce the intake of saturated fat is probably promoting the
consumption of partially hydrogenated plant oils and margarines that are likely to be high in
trans fatty acids. Although the average level of trans fatty acids has declined with the advent
1
of softer margarines, per capita consumption of trans fatty acids has not changed greatly
because of the increased use of commercially baked products and fast foods (Semma, 2002).
In reality, trans fatty acids have been consumed for centuries, since they occur naturally in
beef, mutton, butter, milk and other dairy products (Parodi, 1976). They occur in animal fat
largely because of the microbial hydrogenation in the animal rumen of the polyunsaturated
fatty acids in the foods that animals feed on. Trans fatty acids have also been identified in
very small amounts in some seeds and leafy vegetables (General Conference Nutrition
Council, 2002).
The ingestion of trans fatty acids increases circulating low-density lipoproteins (LDL) to a
degree similar to that of saturated fatty acids, but also reduces high-density lipoproteins
(HDL). Saturated fatty acids do not affect the HDL, therefore trans fatty acids are considered
more atherogenic than saturated fatty acids (Mensink et al., 1990). According to Dr.
Stampfer, Professor of Nutrition at Harvard School of Public Health, trans fatty acids may be
more dangerous to your health than saturated fatty acids. Studies in humans have found that
trans fatty acids are about twice as bad as saturated fatty acids for your blood cholesterol and
triacylglycerol levels (Strampfer, 2004). Mensink et al. (1990) published one of the first
controlled intervention studies that specifically examined the effect of trans mono-unsaturated
fatty acids (MUFA) from hydrogenated plant oils on the serum lipoprotein profile. From this
study, it can be concluded that trans MUFA significantly raises serum total and LDL
cholesterol concentrations and lowers HDL cholesterol, as compared with an iso-energetic
amount of cis MUFA.
The main difference between cis and trans fatty acids isomers is in their geometrical structure.
According to the structure resemblance between trans unsaturated fatty acids and saturated
fatty acids and the differences in structure between cis and trans unsaturated fatty acids, one
can assume that the cis positional isomers are not associated with coronary heart disease
(CHD). These structural differences are demonstrated in Figure 1.
2
Trans unsaturated fatty acid Cis unsaturated fatty acid Saturated fatty acid
Figure 1. The geometrical structure of trans unsaturated, cis unsaturated and saturated
fatty acids
The geometrical structure of the cis isomers are bended, making it difficult to pack together
tightly, while the structure of trans isomers are straight and very similar to that of saturated
fatty acids making it possible to pack together tightly.
Trans fatty acids are well absorbed and it has been estimated that approximately 95% of trans
MUFA are absorbed, which is similar to the rate of absorption of other fatty acids (Emkin,
1979). Other studies have shown that the position of the double bonds of cis and trans
isomers have no effect on the absorption efficiency of these fatty acids (Emkin, 1997). After
absorption, trans fatty acids follow the same metabolic routes as other fatty acids (Emkin,
1984).
Lipoprotein (a) (Lp (a)) concentrations in plasma have been associated with a higher risk for
developing cardiovascular diseases (Lippi et al., 1999). After consumption of a meal high in
trans fatty acids, blood Lp (a) concentrations have been reported to be increased in a number
of publications (Lichtenstein et al., 1999; Sundram et al., 1997).
Another effect of trans fatty acid intake was published for the first time in 1961 by Anderson
and his group. They noted that partially hydrogenated corn oil resulted in higher serum
triglyceride levels than natural oils and butter (Anderson et al., 1961). A raising effect in
3
triglycerides was also seen in a number of recent studies that compared the effect of trans
unsaturated fatty acids with cis unsaturated fatty acids in the blood lipid profile of humans
(Lichtenstein et al., 1999; Sundram et al., 1997). No effect on serum triglyceride levels has
been observed when substituting cis unsaturated fatty acids with saturated fatty acids
(Mensink et al., 1992). Thus, trans fatty acids increase serum triglyceride levels when
compared with other fatty acids.
In addition to influencing lipid risk factors for cardiovascular disease, trans fatty acids have
also been implicated in breast cancer, and in poor faetal development and reduced early infant
growth (Kohlmeier et al., 1997; Koletzko, 1992).
Presently, there is no specific method that permits analysts to distinguish between naturally
occurring trans fatty acids and those produced industrially. This is because of the varying
double-bond positions of trans fatty acid isomers in different hydrogenated oils (Wolff,
1995). Either infrared spectroscopy (IR) or gas liquid chromatography (GLC) is normally
used to identify trans fatty acids in oils and fats. The IR method is not very reliable and lacks
sensitivity for total trans fatty acid content below 5%. IR spectroscopy also does not
distinguish individual trans fatty acids or detect positional isomers (Duchateua et al., 1996;
Ulberth et al., 1996; Firestone et al., 1965). GLC can quantify the trans fatty acid contents as
low as 0.01%, as well as identify some fatty acid isomers, assuming the analysts are well
seasoned and using the latest available technology (Tang, 2002). The development of very
long capillary columns coated with highly polar stationary phases has made it possible to
separate some cis and trans fatty acid isomers (Christie, 1989). However, complete separation
of all the trans and cis isomers is still very difficult with GLC analyses alone, as some
isomers overlap. Identification can be improved by thin layer chromatography on silver nitrate
impregnated silica plates (Ag-TLC) followed by GLC, but this method is very laborious
(Christie, 1989; Molkentin et al., 1995). Good separations have been reported using long
highly polar columns and the optimisation of the GLC oven temperature (Duchateua et al.,
1996).
Complete separation and quantification of all the cis and trans isomers are necessary to
provide accurate estimates of all the different fatty acid isomers in foods. This would allow
the identification of the source of trans fatty acids in processed foods and mixed diets, as well
4
as the possible effects that different positional and geometrical isomers can have on diseases.
Elaidic acid (trans-9, 18:1) is the major man-made trans fatty acid found in partially
hydrogenated plant oils and processed foods, while vaccenic acid (trans-11, 18:1) occurs
naturally in foods from animal sources. Because of their differences (specifically, the position
of the double bond), they have very different physiological and biological effects on humans.
(Belury, 2002). Mahfouz et al. (1984), found that feeding hydrogenated fat to animals
decreases the conversion rate of linoleic acid to arachidonic acid because of the inhibitory
effect that some trans fatty acid isomers have on delta 5 and delta 6 desaturase. In this study,
the position of the trans double bond is shown to play a critical role in the degree of inhibition
(Mahfouz et al., 1984). The identification of the position of the trans double bonds is also
relevant because it has been suggested that trans fatty acids from dairy products, which have
different positional trans double bonds, have different effects on the risk of CHD than those
trans fatty acids from partially hydrogenated oils (Willet et al., 1995).
1.2 Aims of the study
The aims of the study are,
(1) To standardise and optimise an analytical technique to identify and quantify the
different cis and trans mono-unsaturated fatty acid isomers in local margarines and
bread spreads by GLC. Many publications have reported on the quantification of the
total concentration of trans fatty acids in food samples, while less work has been done
on the identification and quantification of the different cis and trans unsaturated fatty
acid isomers found in foods made from partially hydrogenated oils.
(2) To evaluate the GLC results by using Capillary Electrophoresis (CE) for the analyses
of the same samples on a comparative basis.
5
1.2.1 Specific objectives
To determine the best sample extraction method to use in the analyses of commercially
available margarines and spreads.
• To determine the best transmethylation solution for use in preparing fatty acid methyl
esters (FAME) of the extracted samples for GLC analyses.
To compare two different GLC column lengths for the identification and quantification of the
different cis and trans fatty acid isomers.
• To optimise the GLC conditions to have a robust method.
• To identify and quantify as many cis and trans fatty acid isomers as possible.
• To use Ag-TLC for the separation of the cis and trans mono-unsaturated fatty acid
fractions before GLC analyses, and to compare the results obtained with those
obtained without Ag-TLC separation.
• To optimise CE conditions and develop a CE method that is faster.
6
CHAPTER 2
LITERATURE REVIEW
2.1 Introduction
This study deals with the methodology to identify and quantify the different cis and trans fatty
acid isomers in partially hydrogenated plant oils. The negative effects of some of the isomers
on the health of humans are well known. From an analytical chemist’s viewpoint, it is
important to know the differences between commercially produced cis and trans fatty acids
isomers and the natural occurring isomers, and how these can be identified and quantified.
What analytical methodologies are available and how can these be improved?
Work published thus far shows that the different positional and geometrical trans isomers
have different effects on the health of humans (Mahfouz et al., 1984). It is a known fact that
not all trans fatty acid isomers have a negative effect on the health of the population (Belury,
2002). For these reasons, an analytical technique using instrumentation that is normally
available in a lipid analytical laboratory was researched to develop a technique to identify and
quantify the different cis and trans isomers, as well as the other fatty acids that normally
occur in partially hydrogenated oils.
2.2 Fatty acids
Fatty acids are a large and diverse group of naturally occurring organic compounds that are
soluble in non-polar organic solvents (e.g., chloroform, ether, acetone and benzene) and
generally insoluble in water. Fatty acids with up to six carbon atoms are considered short-
chain fatty acids. They are more soluble in water than the longer chain fatty acids, and are
therefore more easily digested and absorbed. Furthermore, they do not behave physiologically
like the longer chain fatty acids, since they are more rapidly digested and absorbed in the
intestinal tract. Biochemically, they are more closely related to carbohydrates than to fats.
Fatty acids with eight to ten carbon atoms are said to have a medium chain. As for short-chain
7
fatty acids, studies have shown that intake of these medium-chain fatty acids may result in
increased energy expenditure via fast digestion. They are further known to facilitate weight
control when included in a diet as replacement for long-chain fatty acids (St-Onge et al.,
2002). Fatty acids with 14 and more carbon atoms are considered as long-chain fatty acids.
The building blocks of most lipids are the fatty acids, which are essential for normal cell
functioning and to stay healthy. They are composed of a chain of methylene groups with a
carboxyl functional group at one end. The methyl chain is the fatty part, while the carboxyl
group is the acid. Fatty acids can be saturated: all the carbon atoms have the maximum
number of hydrogen atoms attached to them and have a straight-chain structure. Because of
the straight structure of saturated fatty acid molecules, they can be packed tightly together,
making them relatively dense and solid at room temperature. This cannot be altered by
hydrogenation. They can also be unsaturated, with one or more double bond connecting some
of the carbons. In unsaturated fatty acids, some of the carbon atoms miss some of their
hydrogen atoms and thus form a double bond between those carbons missing their hydrogen
atoms. With the formation of the double bond or bonds, a bend or kink is formed in the chain
at these sites. The more double bonds an unsaturated fatty acid has, the more bended the
molecule will be. Because of these bends or kinks, the molecules cannot stack together easily
and stay fluid at room temperature. These are mostly oils. Figure 2 shows the geometrical
structures of fatty acids with different degrees of saturation. Oils with a high percentage of
saturated fatty acids are normally solid at room temperature. Other oils with a high percentage
of mono-unsaturated fatty acids (with one double bond), such as olive oil, will solidify when
cooled in a refrigerator. Polyunsaturated fatty acids, which have two or more double bonds
and therefore more bends in their physical structure, stay fluid even when refrigerated.
8
Saturated fatty acid Mono-unsaturated fatty acid Polyunsaturated fatty acid
Figure 2. Geometrical structures of fatty acids with different degrees of saturation
When plants or animals make unsaturated fatty acids, they mostly make these kinked or
bended forms: also referred to as cis unsaturated fatty acids. Most fatty acids are straight- or
bended-chain compounds, and frequently have an even number of carbon atoms. Chain
lengths can range from two to more than 80 carbon atoms, but commonly from 12 to 24.
Branched-chain fatty acids are less common but are generally of microbial origin. These
branched-chain fatty acids are usually not of any nutritional significance.
The common fatty acids in plant tissue are C16 and C18 with zero to three double bonds in
the cis configuration. These fatty acids are also abundant in animal tissues, together with other
fatty acids with a wider range of chain lengths and up to six cis double bonds separated by
methylene groups. These methylene-interrupted double bonds are also referred to as non-
conjugated double bonds. Figure 3 gives the chemical structure of a non-conjugated
unsaturated fatty acid.
H H H H H H H H H
R- C – C –C = C – C –C = C– C – C – COOH
H H H H H
Figure 3. Non-conjugated polyunsaturated fatty acid structure
9
Polyunsaturated fatty acids can also be conjugated. Conjugated fatty acids do not have a
methylene group between the two double-bonded carbons as can be seen in Figure 4.
H H H H H H H H
R- C – C – C = C – C = C– C – C – COOH
H H H H
Figure 4. Conjugated polyunsaturated fatty acid structure
The most well known conjugated polyunsaturated fatty acids are probably conjugated linoleic
acids (CLA). CLAs are a series of positional and geometrical isomers of linoleic acid (cis-9,
cis-12, 18:2). Because of bacterial hydrogenation of linoleic acid in the animal’s stomach,
some of the double bonds flip over to the trans position and some even move to different
positions on the carbon chain. However, the most distinctive reaction is the formation of
conjugated double bonds. A number of cis-cis, cis-trans, trans-cis, and trans-trans isomers
with the double bonds at various positions along the carbon chain have been identified. The
cis-9, trans-11 isomer is the most abundant natural isomer present in ruminant fat (more than
90% of total CLA) (Christie, 2003). These trans conjugated polyunsaturated fatty acid
isomers are not classified as trans fatty acids by the American Food and Drug Administration
(Department of Health and Human Services, 2003). Another group of natural occurring fatty acids are the omega-3 and omega-6 long-chain
unsaturated fatty acids. The human body needs, but cannot synthesise these fatty acids and
therefore they are called essential fatty acids. Essential fatty acids are very important, for
example for our immune system. Alfa-linolenic acid (18:3, n-3) is the parent fatty acid of the
omega-3 series. This is found in dark green vegetables and soybean oil, and is converted in
the body to eicosapentaenioc acid (EPA) and docosahexaenoic acid (DHA). Marine algae and
plankton also synthesise EPA and DHA, and therefore relatively high concentrations are
found in the oil from fish that feed on algae. EPA and DHA also display several
pharmacological properties, such as inhibition of inflammation, altered lipoprotein
metabolism, inhibition of arteriosclerosis, decrease in blood pressure and inhibition of tumour
growth (Sanders et al., 1997). In clinical trails, fish oil supplements containing EPA and DHA
have also been shown to bring about some symptomatic relief in rheumatoid arthritis, colitis,
psoriasis and Crohn’s disease (Belluzzi et al., 1996).
10
Linoleic acid (18:2, n-6) is the parent fatty acid of the omega-6 series, and is the major fatty
acid in sunflower and corn oils. This is converted to omega-6 fatty acids, mainly arachidonic
acid (20:4, n-6) in the body. Arachidonic acid is also found in egg yolk and organ meats.
Arachidonic acid can be converted into eicosanoids like prostaglandins, thromboxanes and
leukotriens that are involved in the regulation of the actions of many cells. Most people eating
a western diet that incorporates soft margarines and plant oils, such as sunflower, corn and
peanut oil, get plenty of omega-6 fatty acids in their diets.
One of the main objectives of this study is to identify and quantify another group of fatty
acids, namely, trans mono-unsaturated fatty acids.
2.3 Trans fatty acids
2.3.1 Natural occurring trans fatty acids
Most unsaturated fatty acids in nature have their double bonds in the cis configuration
(Semma, 2002). Certain bacteria can covert these cis unsaturated fatty acids into unsaturated
fatty acids with the double bonds in the trans configuration. In this configuration, some of the
hydrogen atoms are on the opposite side of the double-bonded carbon atoms. This occurs in
ruminants (cows, sheep and goats) where bacterial fermentation in the fore stomach causes
the formation of trans unsaturated fatty acids. These isomers are found in the body fat of
ruminants and in cows’ milk and products, such as butter (Kepler et al., 1966; Mackie et al.,
1991; Hay et al., 1970). Trans fatty acids which occur naturally in beef and dairy products
have very different physiological and biological functions compared to man-made trans fatty
acids that are found in processed foods (Belury, 2002). Data from the Nurses Health study
reveal that while man-made trans fatty acids increase the risk of CHD, naturally occurring
trans fatty acids of animal origin does not increase this risk (Willet et al., 1993).
Naturally occurring trans fatty acids have also been detected in the membrane lipids of
various aerobic bacteria. Bacteria-degrading pollutants, such as Pseudomonas putida, are able
to synthesise these compounds. They are synthesised by a direct isomerisation of the cis
double bond without a shift in the position. This conversion changes the membrane fluidity in
response to environmental stimuli (Keweloh et al., 1996).
11
Lamberto et al. have also identified two unusual trans fatty acids in seaweed that is grown in
natural seawater. They identified trans-3, hexadecenoic acid (trans-3, 16:1) that has been
known to occur as a component of plant photosynthetic lipids, and a novel trans-3,
tetradecenoic acid (trans-3, 14:1) (Lamberto et al., 1994).
The most well-known group of natural occurring trans polyunsaturated fatty acids are
probably conjugated linoleic acid (CLA). CLA is a collective term for a mixture of positional
and geometrical isomers of linoleic acid, in which the two double bonds are conjugated. Dairy
products are rich in CLA. Unlike the non-conjugated trans polyunsaturated fatty acids, CLA
is recognised as possessing health benefits (Scimeca et al., 2000). This has been found to
contain both antiatherogenic (Nicolosi et al., 1997) and anticarcinogenic properties (Ip et al.,
1994). Micro-organisms in the rumen of animal converts cis-9, cis-12, octadecadienoic acid to
mostly cis-9, trans-11, octadecadienoic acid and trans-10, cis-12, octadecadienoic acid,
isomers. These two isomers are the most well-known trans conjugated polyunsaturated fatty
acids (Parodi, 1997). These trans conjugated unsaturated fatty acids are not classified the
same as the non-conjugated trans isomers that are formed during partial hydrogenation of
plant and fish oils. Although these isomers include a trans configuration, they are not true
trans fatty acids according to the definition of the American Food and Drug Administration,
which defines trans fatty acids as ”unsaturated fatty acids that contain non-conjugated double
bonds in a trans configuration”. (Department of Health and Human Services, 2003).
2.3.2 Commercially produced trans fatty acids
Except for the few cases described in the previous section, all other trans fatty acids are man-
made. Wherever there is a double bond in a fatty acid chain, there is a possibility for the
formation of both positional and/or geometrical isomers. With partial hydrogenation, a double
bond may change from a cis position to a trans position (geometric isomerisation) or move to
another position in the carbon chain (positional isomerisation) and both types of isomerisation
may occur in the same molecule (Dutton, 1997).
Saturated fatty acids have a chain of carbon atoms joined by single bonds, allowing for
rotation about the bonds. Naturally occurring unsaturated fatty acids contain double bonds of
a particular configuration, referred to as cis unsaturated fatty acids. The double bond or bonds
12
restrict rotation. With partial hydrogenation some of these cis double bonds are converted to
the trans isomer. Because the double bond restricts rotation, an unsaturated fatty acid can
exist in two forms. The one is the cis form that has two parts of the carbon chain bended
towards each other, with the hydrogen of the double bond on the same side of the chain
(indicated by the two arrows in Figure 5). The other is the trans form that has two parts of the
chain almost linear and with the two hydrogen atoms at the double bond on opposite sides of
the chain (indicated by the two arrows in Figure 5).
Cis unsaturated fatty acid Trans unsaturated fatty acid
Figure 5. The geometrical structures of the cis and trans mono-unsaturated fatty acids
13
The bended configuration of cis unsaturated fatty acids are the result of polarisation of the
hydrogen atoms causing these to repel each other to form this bended chain. These bended
configurations effectively prohibit the fatty acid molecules from packing tightly together. This
means the bonds between the different cis molecules are weaker, resulting in the fat being
either semi-solid with a low melting point or oil. Highly unsaturated vegetable oils are not
suitable for many applications, such as margarines, shortenings and confectionary fats. The
unsaturated oils are thus hardened by catalytic hydrogenation during which the naturally
occurring cis unsaturated fatty acids are partly converted to the unnatural trans isomers.
Depending on the type of unsaturated oil used and the temperature, pressure and duration of
hydrogenation, different trans isomers can be formed. During hydrogenation, a few things can
happen to the unsaturated oil. All the double bonds can be removed to form saturated fatty
acids, or only some of the double bonds can be removed to change polyunsaturated fatty acids
into monounsaturated fatty acids. Some of the double bonds may remain, but be moved in
their positions on the carbon chain. Some of the cis double bonds can be changed into the
trans position to produce several geometrical and positional isomers (Almendingen et al.,
1995).
Trans fatty acids are well absorbed and incorporated into tissue lipids (Emken, 1995) and
similarly transported to other fatty acids to be distributed within the cholesterol ester,
triacylglycerol, and phospholipid fractions of the lipoproteins (Vidgren et al., 1998). The
ingestion of trans unsaturated fatty acids increase low-density lipoproteins (LDL) to a similar
degree of that of saturated fatty acids, but also reduces high-density lipoproteins (HDL).
Therefore, trans fatty acids are considered to be more harmful than saturated fatty acids.
(Ascherio et al., 1997). From the literature it is clear that the trans isomers of mono-
unsaturated fatty acids are causing the negative attitude, rather than the cis positional isomers
(Technical Committee of the Institute of Shortening and Edible Oils, 2006).
2.4 Hydrogenation
In the early 1900s, the only fat available for commercial use was lard, which was rendered
easily and cheaply from pork fat. Lard has a good shelf live and excellent shortening
properties, but is high in cholesterol and saturated fatty acids. With the growing health
concerns about the dangers of eating too much saturated fat, there was pressure to find an
14
alternative, more unsaturated, source. During 1912, French chemist Paul Sabatier won the
Nobel Prize for developing the hydrogenation process. This method allows oil refiners to
modify unsaturated liquid oils to be suitable substitutes for lard (Paterson, 1996). A common
misunderstanding is that partial hydrogenation changes all unsaturated fats to saturated fats.
This is true for total hydrogenation, but with partial hydrogenation only some molecules of
unsaturated fatty acids are converted to saturated fatty acids, while a large percentage of the
natural occurring cis unsaturated fatty acids are converted to trans unsaturated fatty acid
isomers.
Food manufacturers discovered that bubbling hydrogen through unsaturated oils, created
partially hydrogenated fats that have a higher melting point and are less vulnerable to
becoming rancid than the original oils and therefore have a longer shelf life. This process
converts some of the cis or bended forms to a straightened or trans form. The chemical
structure of the two forms is the same. It has the same number of carbon, oxygen and
hydrogen atoms, and the double bond can be between the same two carbon atoms, but with a
different geometrical configuration and it is a straight instead of kinked molecule because of
the trans configuration. The body recognises the double bond and tries to use it for the same
purposes that it uses the cis form, but the trans form stacks together just like saturated fatty
acids, which sabotages the flexible and porous functionality of the cell membranes (Oslund-
Lingvist et al., 1985).
Consider the differences between total hydrogenation and partial hydrogenation. If cis-9, cis-
12, octadecadienoic (18:2), an unsaturated fatty acid with two double bonds in the cis
positions and a melting point of -7oC, is 100% hydrogenated, the two double bonds will be
forced to break to form single bonds. An additional four hydrogen atoms will be added to the
molecule. As a result of the total hydrogenation process, the bended molecule becomes a
straight chain and the melting point of the oil will be changed to 70oC and the structural
configuration will resemble that of stearic acid (18:0). For all practical purposes this is a
stearic acid. However, besides being costly it takes much energy to produce a saturated fatty
acid that is naturally occurring and is also just too hard a fat to be made into margarine and
shortening. Depending on the melting point of the fat that you need, you can partially
hydrogenate the original oil to produce unsaturated oil with a specific melting point. For
example, the partial hydrogenation of cis-9, cis-12, octadecadienoic fatty acids can produce
15
several different geometrical and positional isomers; only one double bond can break to give
you cis-9, octadecenoic acid (18:1). It is still a bended molecule, but not as much as the
original molecule, and its melting point is increased to 16oC, or can change to the trans-9,
octadecenoic acid isomer with a melting point of 44oC and a straight geometrical structure.
During the partial hydrogenation process, the double bond can even change to a different
position on the carbon chain, for example to position 11 to give you cis-11, octadecenoic acid
with a melting point of 12oC and still be a bended structure, or to a trans-11, octadecenoic
acid with a melting point of 39oC, but again with a straight structure. All the different isomers
of octadecenoic acid have the same molecular weight. While it is still an unsaturated fatty
acid, it clearly is the number of double bonds as well as the positional and geometrical
structure that determines the melting point of the final product. Vegetable oil is too soft to
make margarine and shortening, and saturated fat is too hard. An in-between product is
needed which is why the industry only partially hydrogenates the vegetable oils.
During the partial hydrogenation process, which is easily controlled, hydrogen atoms are
added in no particular order. When the hydrogenation process is stopped, unsaturated fatty
acids are in varying stages of hydrogenation. Some molecules are totally hydrogenated
(saturated) while in others, some of the double bonds have changed from the natural cis
configuration to the unnatural trans configuration. Some of the double bonds have even
shifted to unnatural positions on the carbon chain. During the partial hydrogenation process,
the bent cis isomer changes to the trans isomer forming a molecule that has a straight
configuration, similar to saturated fatty acids. The straight configuration of trans unsaturated
fatty acids enable the molecules to pack easily together resulting in a higher melting point
with a longer shelf life and flavour stability. These more stable fats are used in margarines and
shortenings.
2.5 Naming of fatty acids (Nomenclature)
Fatty acids are normally classified into two groups, either saturated or unsaturated.
Unsaturated fatty acids can further be classified into monounsaturated, with one double bond
or polyunsaturated with two or more double bonds. The unsaturated fatty acids derive their
systematic names from the parent unsaturated hydrocarbon. The unsaturated fatty acid,
octadecenoic acid (18:1) is derived from the hydrocarbon octadecene. The number of double
16
bonds in a polyunsaturated fatty acid chain is designated by the terms di-, tri-, tetra-, etc.,
inserted into the name as in octadecadienoic acid (18:2) a polyunsaturated fatty acid with two
double bonds and octadecatrienoic acid (18:3), a polyunsaturated fatty acid with three double
bonds. (Perkin, 1991). The numeric designations used for fatty acids come from the number
of carbon atoms followed by the number of double bonds. To precisely describe the structure
of a fatty acid molecule, the length of the carbon chain (number of carbons), the number of
double bonds and also the exact positions of these double bonds must be known. This will
define the biological reactivity of the fatty acid molecule.
According to official International Union of Pure and Applied Chemistry (IUPAC)
nomenclature, the carbons in a fatty acid chain are numbered consecutively with the carbon of
the carboxyl group being considered number one. This is also the form of nomenclature
preferred by the International Commission on Biochemical Nomenclature. By convention, the
lower number of two carbons that have a double bond identifies the first double bond in a
chain. For example, in cis-9, octadecenoic acid (18:1) the double bond is between the 9th and
the 10th carbon atom and it is a cis isomer. Another form of nomenclature designates
octadecenoic acid as 18:1(n-9), which indicates that the double bond is 9 carbons away from
the methyl group. Although this contradicts the convention that the position of the double
bond should be counted from the carboxyl end of the carbon chain, it is of great convenience
to lipid biochemists, because the number of the last double bond remains the same when
carbon atoms are added or removed from the carboxyl end during metabolism (Nutritiondata,
2006).
The polyunsaturated fatty acid, cis-9, cis-12, octadecadienoic acid (linoleic acid), explains the
nomenclature better. Counting the carbon atoms from the carboxyl group, the first double
bond is between the 9th and the 10th carbon and the second double bond is between the 12th
and the 13th carbon, and the hydrogen atoms on the carbon atoms, at both double bonds, are in
the cis positions. Counting the carbon atoms from the methyl group the first double bond is
between the 6th and the 7th carbon. This is why linoleic acid is also known as 18:2 (n-6), an
omega-6 polyunsaturated fatty acid (Figure 6).
17
Figure 6. Geometrical structure of cis-9, cis-12, octadecadienoic acid, also known as
18:2 (n-6)
The aim of this study is to identify the different cis and trans fatty acids isomers. Therefore,
the nomenclature as preferred by the IUPAC is the most appropriate. With this system, all the
different positional and geometrical fatty acid isomers can be identified by their names.
The following list (Table 1) gives the scientific names, shorthand designation and the trivial
name of some of the fatty acids used in this thesis.
These are just a few of the most common fatty acids. With partial hydrogenation, the natural
monounsaturated and polyunsaturated fatty acids can form a number of different positional
and geometrical fatty acid isomers each with their own name.
18
Table 1. The scientific names, shorthand designation and trivial names of some of the fatty acids
Saturated fatty acids
Scientific name Shorthand designation Trivial name
Dodecanoic acid 12:0 Lauric acid
Tetradecanoic acid 14:0 Myristic acid
Hexadecanoic acid 16:0 Palmitic acid
Heptadecanoic acid 17:0
Octadecanoic acid 18:0 Stearic acid
Monounsaturated fatty acids
Shorthand designation Trivial name Scientific name
Cis-9, Tetradecenoic acid Myristoleic acid 9-14:1
Cis-9, Hexadecenoic acid 9-16:1 Palmitoleic acid
Trans-9, Hexadecenoic acid 9-16:1 Palmitelaidic acid
Cis-6, Octadecenoic acid 6-18:1 Petroselinic acid
Cis-9, Octadecenoic acid 9-18:1 Oleic acid
Cis-11, Octadecenoic acid 1-18:1 Vaccenic acid
Trans-6, Octadecenoic acid 6-18:1 Petroselaidic acid
Trans-9, Octadecenoic acid 9-18:1 Elaidic acid
Trans-11, Octadecenoic acid 11-18:1 Trans-vaccenic acid
Polyunsaturated fatty acids Scientific name Shorthand designation Trivial name
Cis-9,Cis-12, Octadecadienoic acid 9c,12c-18:2 Linoleic acid
Cis-9,Trans-11, Octadecadienoic acid 9c,11t-18:2 Conjugated linoleic acid
19
2.6 Analytical procedures for the determination of cis and trans fatty acids
2.6.1 Introduction
Currently there are two official methods for the quantification of trans fatty acids as accepted
by the American Oil Chemists’ Society (AOCS) and the Association of Official Analytical
Chemists (AOAC), namely GLC and IR Spectroscopy. Several other analytical methods are
reported for trans fatty acid determination and quantification in food. These analytical
procedures mostly stem from separative techniques generally used for lipid analyses namely,
Ag-TLC, high performance liquid chromatography (HPLC), high performance liquid
chromatography with packed columns impregnated with silver nitrate (Ag-HPLC) as well as
the two accepted methods. Each of these methods has advantages and drawbacks.
Improvements in the accuracy and effectiveness of the results can be obtained by combining
some of these methods.
2.6.2 Infrared spectroscopy
Infrared spectroscopy is the method that was used over the last few decades to determine the
total trans fatty acid composition of food samples. Trans ethylenic bonds show a specific
absorption in the infrared spectrum at 967 cm-1. This method is fast and easy for routine
analyses, but the IR method is not very reliable and lacks sensitivity for total trans fatty acid
content below 5%. IR spectroscopy also does not distinguish individual trans fatty acid
isomers or detect positional isomers (Duchateua et al., 1996). Furthermore, results obtained
using IR spectroscopy is higher, sometimes as much as twice those obtained by GLC
(Ulbrecht et al., 1994). Several reasons could explain these discrepancies. Most
triacylglycerols are absorbed in the infrared spectrum at a similar wavelength as the trans
isomers, which lead to an apparent increase in the trans fatty acid level measurements
(Deman et al., 1983). IR spectroscopy also measures conjugated trans fatty acid isomers,
which are not considered real trans fatty acids (Ulbrecht et al., 1994). Limitations in the use
of IR spectroscopy to determine the trans fatty acids content of food samples were the lack of
accuracy, especially at low levels of trans fatty acid isomer content, and the inability to
distinguish between the different positional and geometrical isomers (Ulbrecht et al., 1996;
Firestone et al., 1965). Emergence of Fourier-transform infrared spectroscopy (FTIR) and the
20
use of computer-assisted spectral subtraction procedures allowed for the improved detection
efficiency of this method. Unfortunately, this method still yielded somewhat higher levels
than the values recorded when using GLC. Furthermore, some large variations were noticed in
the measurement of oils that contained low levels of trans fatty acids, as usually is the case
with partially hydrogenated oils (Ulbrecht et al., 1994).
2.6.3 Silver impregnated thin layer chromatography
Geometric isomer separation using Ag-TLC is based on the property of trans isomers, which
form unstable compounds in reaction to silver salts. These compounds are different from
those formed with cis isomers (Ledoux et al., 2000). In most cases, the thin layer plates were
dipped in a 5-20% silver nitrate solution, then dried and activated. The fatty acid methyl ester
samples were then spotted and developed in saturated tanks in hexane-diethyl ether or
petroleum ether- diethyl ether. This led to the separation of the cis and trans monounsaturated
fatty acid fractions. The cis and trans monounsaturated fatty acid methyl ester spots were then
scraped off the silica gel plates and analysed by GLC (Precht et al., 1997). GLC analyses after
Ag-TLC, led to much better results than the use of GLC alone. Molkentin et al. (1995)
succeeded in separating 10 peaks for trans 18:1 fatty acids and 9 peaks for cis 18:1 isomers
using a 100 m CP Sil-88 capillary column after pre-separation by Ag-TLC. Ledoux and his
group (2000) obtained 18 different peaks using similar operating conditions. This method has
the drawback of being very time-consuming and laborious, with no possibility of automation.
On the other hand, it is a cheap and easy method to use.
2.6.4 High performance liquid chromatography
The use of HPLC for the identification and quantification of different cis and trans fatty acid
isomers is one of the newer methods. Juanèda (2002) published a paper on the use of a HPLC
fitted with two reverse-phase columns for the separation of the cis and trans isomers, but
GLC still had to be used to analyse the collected fractions. In this study, an expensive HPLC
was used only to separate the cis and trans isomers, while a GLC was still needed for the
identification and quantification of the different isomers. The appearance of commercial
silver-ion columns for HPLC has caused a revival of this technique. A number of papers have
been published on the use of silver-ion high-performance liquid chromatography to identify
21
isomeric cis and trans fatty acids over the past few years (Adlof, 1994, Ratnayake, 2004).
Both the capital and running costs of this technique are much higher than that of GLC. The
complex nature of the separation process causes the identification of compounds emerging
from HPLC to be complicated (Christie, 1989).
2.6.5 Gas liquid chromatography
Fatty acids are the group of lipids most commonly analysed by GLC. It is undoubtedly the
technique that would be mostly chosen for this purpose (Stoffel et al., 1959). The major
advances in this method, regarding the identification and quantification of the different cis
and trans fatty acids isomers, are the commercial availability of very long capillary columns
packed with highly polar stationary phases. Column efficiency is proportional to the square
root of column length, and resolution is influenced by the selectivity of the stationary phase.
Increasing column length will therefore lead to higher resolution, and modification of the
stationary phase will effect separation (Wolff et al., 1995).
Recently available highly polar columns bonded with cyanoalkyl polysiloxan phases, such as
SP-2560 (Thompson, 1997) and BPX-70 (Berdeaux et al., 1998), demonstrated significant
improvements in the separation and quantification of the different cis and trans isomers. By
using cyanoalkyl polysiloxan as a stationary phase, trans 18:1 isomers are eluting in the
double-bond position progression along the carbon chain from the carboxylic acid end of the
fatty acid chain (trans-4, trans-5, trans-6, trans-7…). Most of the trans isomers also have
shorter retention times than those of oleic acid (cis-9, 18:1) (Aro et al., 1998). Quantitation of
the main trans isomer in milk fat, vaccenic acid (trans-11, 18:1) (Molkentin et al., 1995),
together with trans-9, 18:1 and trans-10, 18:1, which represent the major trans isomers in
hydrogenated plant oils (Parodi, 1976), can easily be done with these columns (Aro et al.,
1998). From these chromatograms the source of the trans isomers in processed foods, can be
identified. The superb resolutions attainable with the new very long highly polar, wall-coated
open tubular (WCOT) capillary columns, make it more challenging to use GLC fitted with
these columns for the identification and quantification of the different cis and trans fatty acid
isomers in partially hydrogenated oil samples.
22
2.6.6 Capillary electrophoresis
Capillary electrophoresis is a highly efficient and flexible analytical separation technique that
has become a serious competitor for GLC, but a number of problems remain to be solved. A
very small sample volume is required for CE analyses. This can negatively impact precision
and sensitivity. More importantly though, is the degree to which the small volume is
representative of the overall sample, since it remains very problematic especially when
working with oils with a low percentage of trans fatty acids (Castaneda et al., 2005). One way
of minimising these problems and their strong effect on the quality of the results is to consider
sample preparation as a key part of CE processes. (Valcarcel et al., 1998).
Fatty acids are normally analysed by GLC, but there is still a need to speed up the analytical
time. An attractive alternative separation technique may possibly be CE, and in particular,
micellar electrokinetic chromatography (MEKC) (Erim et al., 1995). An advantage of MEKC
is the fact that compounds that are insoluble in aqueous solutions, like fatty acids, can be
solubilised. The absence of a chromophoric or fluorophoric group in fatty acids excludes
direct UV detection and therefore indirect detection has to be used (Erim et al., 1995). So far
most of the articles on fatty acids with CE dealt with the analyses of saturated and unsaturated
short, medium- and long-chain fatty acids. There were a few publications on trans fatty acids,
but they dealt mostly with the identification of cis and trans isomeric groups and not so much
on the identification of the different isomers. An article by de Oliveira et al. (2003) described
a method to analyse trans fatty acids in hydrogenated oils by CE. They used indirect UV
detection with sodium dodecyl benzenesulfonate as a chromophore and a neutral surfactant,
polyethylene 23 lauryl ether. Elaidic acid (trans-9, 18:1) and oleic acid (cis-9, 18:1), as well
as other saturated and unsaturated fatty acids were separated in hydrogenated Brazil nut oil
(de Oliveira et al, 2003).
2.7 Lipid extractions
Very few papers deal with lipid extraction in depth, yet the correct extraction procedure is the
first critical step in the identification and quantification of fatty acids. Quantitative isolation of
all the lipids in the sample in their native state and which are free of contaminants must be
accomplished before being analysed. Care must be taken to minimise the risk of hydrolyses
23
and oxidation of the fatty acids. To extract the fatty acids, it is necessary to find solvents that
will not only dissolve the lipids readily, but will also overcome the interaction between the
lipids and the sample matrix. Most lipid analysts use a mixture of chloroform and methanol to
extract the lipids from animal and plant material (Christie, 1993). Over the years some interest
has been shown in iso-propanol/hexane (2:3), because its toxicity is relatively low, but much
more testing needs to be done on its extraction ability (Radin, 1981). Benzene was also
frequently mentioned as a solvent with very good extraction properties. However, today this is
known to be extremely toxic and other solvents are preferred, even though most solvents
exhibit some degree of toxicity if inhaled.
Margarines consist mainly of triacylglycerol molecules with very little non-lipid
contaminants, making the extraction procedure quite simple. Any lipid lacking polar groups,
for example triacylglycerols, are soluble in moderately polar solvents such as chloroform, and
very soluble in hydrocarbons such as hexane (Christie, 1993). Most of the literature
describing the extraction of lipids from fat and oils mention the use of a mixture of
chloroform and methanol that is based on the method first published by Folch et al. (1957).
Bligh et al. (1959) published a simple adaptation of the original Folch method merely as an
economical means of extracting lipids from fish. Others tried the Bligh method and found it
lacking in the recovery of non-polar lipids (Cabrini et al., 1992). Richardson et al. (1997) also
described a modification of Folch’s extraction method that gave excellent results. These
scientists used large volumes of chloroform and methanol, with a final ratio of 1:1(v:v), to
extract the fatty acids and then used a rotary evaporator to remove the solvent (Richardson et
al, 1997). Lepage et al. (1984) used a method where they left out the extraction step and
directly transmethylated the samples with good results. Some work was also published on the
combination of the extraction and transmethylation steps (Kang et al., 2005).
It is obvious that no matter what extraction procedure is used, great care should be taken to
guard against oxidation of the extracted fatty acids. The use of an antioxidant such as
butylated hydroxytoluene (BHT) must always form an integral part of this procedure. Where
possible, fatty acid extracts should also be handled in an atmosphere of nitrogen (Christie,
1993).
24
A mixture of chloroform and methanol is probably the best general lipid extraction solution,
but it is not the safest from an environmental standpoint and n-hexane is an extraction solution
worth trying (Christie, 1993).
2.8 Transmethylation of fatty acids
Before the fatty acid components of any lipid can be analysed by GLC, it is necessary to
convert them to low molecular weight non-polar derivatives, such as methyl esters. Although
fatty acids can occur as free fatty acids in nature, they are mostly found as esters linked to a
glycerol. Nearly all the important fats and oils of animal and plant origin consist almost
exclusively of this simple lipid class, and are known as triacylglycerols (Figure 7) or
commonly as triglycerides (Christie, 1989).
Figure 7. The chemical structure of a triacylglycerol molecule
The preparation of methyl esters derivatives from triacylglycerols is by far the most common
chemical preparation performed by lipid analysts. In short, it means the breaking of the bond
(hydrolyses) between the fatty acids and the glycerol backbone and the formation of a fatty
acid methyl ester. There is no need to hydrolyse or saponify triacylglycerols to obtain free
fatty acids before preparing the methyl esters, as they can be transesterified or
transmethylated directly to fatty acid methyl esters (FAME) for GLC analyses (Christie,
1990). A number of different transmethylation methods have been described to form
derivatives, depending on the samples and methods the analysts were using. Since this study
25
focuses on the identification and quantification of cis and trans fatty acids isomers in partially
hydrogenated oils, only the two most common transmethylation methods for triacylglycerols
will be reviewed.
2.8.1 Acid-catalysed transmethylation
Classic acid catalysed transesterification chemistry calls for the reaction of a triacylglycerol
with alcohol in the presence of an acid catalyst to form FAME (Figure 8) for GLC analyses.
RCOOR’ + CH3OH RCOOCH3 + R’OHH+1 2 3 4 5
RCOOR’ + CH3OH RCOOCH3 + R’OHH+1 2 3 4 5
Figure 8. Acid-catalysed transmethylation reaction of a lipid to form a methyl ester 1. Lipid 2. Methanol 3. Acid 4. Methyl ester 5. Glyserol
A commonly used acid transesterification catalyst is probably boron trifluoride (BF3) in
methanol. This reagent could be used to transmethylate most lipid classes (Morrison et al.,
1964). Although this reagent has serious drawbacks, it has the advantage that it can be easily
purchased from a number of suppliers (Christie, 1994). By using BF3 in methanol, methoxy
artifacts are produced from unsaturated fatty acids by adding methanol across the double bond
when high concentrations of this reagent are used (Lough, 1964). There is some evidence that
the artifact formation is most likely the result of aged reagents (Fulk et al., 1970). The reagent
has a very limited shelf life at room temperature, and if it has to be used, this should carefully
be checked beforehand.
The most frequently cited reagent for the preparation of methyl esters is 5% anhydrous
hydrogen chloride in methanol (Christie, 1990). In the standard transmethylation procedure,
the samples were dissolved in at least a 100-fold excess of methanolic hydrogen chloride and
refluxed for 2 hours or held at 50oC overnight. Triacylglycerols are not soluble in methanolic
hydrogen chloride alone, and an inert solvent, like hexane or chloroform, must be added to
affect the solution before the reaction can be started. After the reaction, the methyl esters are
extracted by adding water and hexane. All the different fatty acids are esterified at
26
approximately the same rate, so there is unlikely to be differential losses of specific fatty acids
during the methylation step, except for short-chain FAME that may be lost during refluxing of
the sample. Short-chain FAMEs are also soluble in water and can be lost during an aqueous
extraction step. The best transmethylation methods for short-chain fatty acids are those that
require no heat or aqueous extraction steps (Christopherson et al., 1969).
Hydrogen chloride in methanol can be said to be the best general purpose esterifying reagent
available, but its long reaction time is a disadvantage. A good alternative to hydrogen chloride
in methanol is a 5% solution of concentrated sulphuric acid in methanol. This is very easy to
prepare and is the preferred reagent for transmethylation of triacylglycerols (Christie, 1990).
As sulphuric acid is a very strong oxidising reagent, great care must be taken when working
with polyunsaturated fatty acids. However, there is no evidence of side effects when using a
5% sulphuric acid solution and moderate temperature for a short time (Christie, 1990).
2.8.2 Base-catalysed transmethylation
Triacylglycerols are transmethylated very rapidly in anhydrous methanol in the presence of a
basic catalyst such as sodium methoxide, which facilitates the exchange between glycerol and
methanol (Figure 9). The reaction is very quick and triacylglycerols are completely
transesterified at room temperature in a few minutes (Marinetti, 1966).
RCOOR’ + CH3OH RCOOCH3 + R’OH-OCH3
1 2 3 4 5
RCOOR’ + CH3OH RCOOCH3 + R’OH-OCH3
1 2 3 4 5-OCH3
1 2 3 4 5
Figure 9. Base-catalysed transmethylation reaction of a lipid to form a methyl ester 1. Lipid 2. Methanol 3. Base 4. Methyl ester 5. Glycerol
The most popular basic transmethylation solvent in use is a 0.5 to 2 M sodium methoxide in
anhydrous methanol. This mixture is stable for several months at refrigeration temperature if
oxygen-free methanol is used for its preparation (Christie, 1972). As with acid catalysed
transesterification procedures, a further solvent, such as hexane, ether or toluene is needed to
solubilise the non-polar lipids after extraction. In a typical transmethylation reaction, the
sample is extracted and dissolved in sufficient hexane or other solvent to get it into solution.
27
A 100-fold excess of 0.5 M sodium methoxide in anhydrous methanol is then added. After
about 10 minutes, 0.1 ml glacial acetic acid is added to neutralise the sodium methoxide, and
the methyl esters are extracted using water and hexane. Sodium methoxide in anhydrous
methanol is a useful reagent for fast transmethylation of fatty acids linked by ester bonds to
alcohols, but cannot form methyl esters from free fatty acids (Jamieson et al., 1969).
To evaluate which transmethylation method is the most suitable for the formation of FAME
from margarines, two transmethylation reagents will be used: a 5% concentrated sulphuric
acid in double distilled methanol solution, and a 0.5 M methoxide in anhydrous methanol
solution.
2.9 Instrumentation
2.9.1 Gas liquid chromatograph
2.9.1.1 Introduction
In 1901, the Russian botanist, Mikhail Tsvet, invented the first chromatography technique
during his research on chlorophyll. He used liquid-adsorption columns to separate plant
pigments. The technology of chromatography advanced rapidly throughout the 20th century,
and today we have highly sophisticated instrumentation. Chromatography is a separation
method that exploits the difference in partitioning behaviour of different compounds in a
mixture between a mobile and a stationary phase, to separate the compounds. This involves a
sample being carried in a mobile phase, normally a gas, and forced through a stationary phase
that is packed into a column. The components of the sample which are to be separated have
different affinities for the mobile and the stationary phases. Those components which have a
higher affinity for the mobile phase will move faster through the column than those which
have less affinity for the mobile phase, but more for the stationary phase. The components
that have the most affinity for the stationary phase will move the slowest through the column.
As a result of these differences in mobilities, sample components will become separated from
each other as they travel through the column. The various components of a sample elute at
different times (called retention times) resulting in separation and identification (McMurry,
2000). The distinguishing feature of GLC, is that the mobile phase is a gas, and the stationary
28
phase a liquid. The liquid is either coated on the wall of the column or coated onto the surface
of small particles that are fixed to the wall of the column. The composition of the stationary
phase generally has the greatest effect on the separation of the different components of a
sample. This is because the mobile gaseous phase mainly serves as the carrier of the sample.
A detection device at the end of the column records each substance as it is emitted, noting the
length of time and the concentration. The length of time (retention time) identifies the
substance, and the intensity of the response reveals the concentration of the substances in the
sample. These two indicators are plotted on a graph (chromatogram). These chromatograms
are then compared to a chromatogram of a standard mixture that is analysed under the same
conditions. In this manner the components of the sample can be identified (Willet, 1987).
2.9.1.2 Gas liquid chromatograph instrument
The modern gas chromatograph consists of five units (Figure 10).
Figure 10. Basic components of a GLC
Courtesy: http://www.shu.ac.uk/gaschrm.htm
The gas supply unit provides all the required gas provisions to the instrument. Next is the
sampling unit or the injector. Typically, the injector has its own temperature-controlling unit
that monitors and controls the temperature. The oven controls the column temperature and is
the most significant part of the system. The detector is located in its own oven for temperature
regulation. The flame ionisation detector (FID) is the most commonly used detector in
29
modern GLC, because of its simplicity and reliability (Willet, 1987). The GLC system
concludes with a data-processing computer, which drives the system and acquires the
detectors output, processes this and prints the report. The chromatogram obtained from this
procedure is always compared to the chromatogram of a standard sample run through the
same instrument under the same conditions as the unknown sample.
2.9.1.3 Carrier gases
The type of carrier gas is important, and hydrogen is preferred to nitrogen and helium,
because of its low resistance to mass transfer. Column efficiency also varies less with gas
velocity over the useful working range when hydrogen, instead of nitrogen or helium is used,
so that precise flow calibration is less critical (Christie, 1989). The Van Deemter plot (Figure
11) of the variation in the height of an effective theoretical plate illustrates this clearly.
Figure 11. Van Deemter plot indicate the effect of the velocity of nitrogen, helium and
hydrogen as a carrier gas on the theoretical plate height. (The lower the plate height, the better the separation)
When the flow rate of the carrier gas is too low, there is a tendency for band broadening
through longitudinal diffusion. At a too high flow rate, band broadening is diminished, but
there may be insufficient time for the different components of the sample to enter into the
liquid or stationary phase and thus poor separation occurs.
30
2.9.1.4 Columns
Two general types of column are available. These are the packed and capillary columns. This
study focuses on the use of capillary columns. The latest type of capillary column is made
from fused silica with an internal diameter between 0.18 and 0.53 mm. They are known as
wall-coated open tubular (WCOT) columns. WCOT columns consist of an open capillary tube
with walls coated with a liquid stationary phase. Fused silica has proved to be an excellent
medium that consists of an amorphous silicate material that is free of metal oxides and
therefore very inert. They are coated on the outside with a polymeric material to prevent
fractures. In the latest fused silica WCOT columns, the liquid phase is bonded chemically to
the inside surface of the column, and the individual molecules of the polymeric liquid phase
are cross-linked by chemical methods to improve their stability at high temperatures (Christie,
1989). The principal requirement of a liquid phase is to provide the correct degree of
selectivity for the separation of the components in a sample. In lipid analyses, the main factor
influencing separation is the polarity of the liquid phase (Christie, 1989). For the analyses of
the different cis and trans isomers in partially hydrogenated oils, a number of different
columns have been used. Most of these were very long capillary columns coated with highly
polar cyanoalkyl polysiloxane stationary phases, marketed under trade names such as BPX-70
(biscyanopropylsiloxane polysilphenylene), SP-2340 (5% cyanopropyl phenyl polysiloxane +
95% bicyanopropyl polysiloxane) and CP-Sil 88 (bicyanopropyl polysiloxane) (Precht et al.,
1996; Aro et al., 1998; Ball et al., 1993). From the literature it is clear that you need a long
column with a polar stationary phase in the first place, and secondly, most experiments use
isothermal column temperatures between 160o oC and 180 C (Duchateua et al., 1996;
Ratnayake et al., 2002).
2.9.1.5 Column temperature
Optimal separation of the components in the sample is greatly dependant on the column
temperature. The retention time of a compound depends, not only on the type of stationary
phase or the velocity of the carrier gas, but also on the column temperature. A higher column
temperature will cause the compound to move faster through the column giving a shorter
retention time, but this will also cause a decrease in the efficiency of the column. Although
the components emerge as sharper peaks because of the increase of the vapour pressure of the
31
solute, the ratio of the solute in the gas phase to the liquid phase also increases. If the column
temperature is too high, highly volatile components in the sample will move practically as fast
as the carrier gas and will not be separated. On the other hand, at a too low column
temperature, the less volatile components will hardly move through the column. Column
temperature plays a critical role in the optimal separation of the different cis and trans isomers
in partially hydrogenated oils. Ratnayake et al. (2002) found that a column temperature of
180oC produced the fewest overlapping peaks of cis and trans isomers on 100 m SP- 2560
and CP-Sil 88 capillary columns. At this temperature all trans isomers, except trans-13 and -
14, and trans-15, 18:1 isomers were resolved. Isothermal temperatures above and below
180oC produced some additional overlapping (Ratnayake et al., 2002). Aro et al. (1998) also
used a CP-Sil 88 capillary column, but they reported that a column temperature of 155oC gave
the best resolution of the cis and trans isomers.
No golden standard for optimal column temperature exists and each analyst must optimise the
column temperature to suit his or her conditions and applications.
2.9.1.6 Injectors
Samples are applied to the column by means of an injector. With the use of capillary WCOT
columns, different types of injectors can be used. For optimum column efficiency, the sample
should not be too large or too concentrated and should be introduced onto the column as a
plug of vaporised sample. During the injection process, it is important that the sample should
not change in composition and there should be no discrimination against any component in
the sample. Thermal degradation or rearrangement should be negligible. No loss of column
efficiency should be introduced. The solvent peak should not interfere with the detection of
the solutes, and the detection time and peak areas should be reproducible. Capillary WCOT
columns have a very small sample capacity, and it is relatively easy to overload a column by
injecting a too large sample volume or very concentrated sample. The split injector (Figure
12) is ideal to circumvent this problem. In this study only the split injector will be used.
32
Figure 12. Split/ Splitless injector Courtesy: http://www.shu.ac.uk/gaschrm.htm
The injector contains a glass liner into which the sample is injected through a septum. After
injection the sample vaporises because of the high temperature of the injector and forms a
mixture of carrier gas, vaporised solvent and vaporised solutes. The sample, mixed with the
carrier gas, enters the injector chamber and leaves the chamber via three routes: 1) a portion
purges the septum to prevent septum-bled components entering the column, and leaves via the
septum purge outlet, 2) a portion flows on to the column, and 3) the largest portion of the
sample exits through the split outlet.
2.9.1.7 Detectors
There are a large number of detectors that can be used in gas chromatography, but only a few
are used to a significant extent of which the FID is the most popular (Figure 13). The FID is
highly sensitive and stable, and has a low dead volume, a fast response time and is linear over
a very wide concentration range. Only inert gasses and a few other volatile substances do not
give a substantial signal. One great disadvantage of a FID is that the sample is destroyed
because of the use of a hydrogen diffusion flame to ionise the compound for analyses. The
FID responds to any molecule with a carbon-hydrogen bond. Since the FID is mass sensitive
and not concentration sensitive, changes in the carrier gas flow rate have little effect on the
33
response of the detector. As the sample elutes from the column, it is mixed with hydrogen and
passes through a flame that breaks down the molecules and produces ions. These ions carry a
current that is measured by the detector, amplified and sent to the data-processing system.
Figure 13. Schematic drawing of a flame ionisation detector Courtesy: http://www.shu.ac.uk/gaschrm.htm
2.9.2 Capillary electrophoresis
2.9.2.1 Introduction
Electrophoresis is defined as the migration of ions in an electrical field. When a positive
(anode) and a negative (cathode) electrode are placed in a solution containing ions and a
voltage is applied across the electrodes, the anions and the cations in the solution will move
towards the electrode with opposite charge. Separation by electrophoresis relies on the speed
of the mobility of the different ions in the sample. The mobility will be determined by the
charge as well as the size of the ions. The higher the charge and the smaller the ion, the faster
it moves.
34
Another very important feature of capillary electrophoresis is the flow of the buffer liquid
through the capillary column that is normally made from fused silica. The surface of the
inside wall of the fused silica capillary is made up of ionisable silanol groups which dissociate
to produce anions (SiO-), especially above a pH of 4. These groups give the wall a negative
charge. When the capillary is filled with a buffer solution, the negatively charged wall will
attract the positive ions in the buffer solution creating an electrical double layer and a
potential difference (zeta potential) close to the capillary wall (Figure 14).
Figure 14. Stern’s model of the double-layer charge distribution at a negatively charged capillary wall leading to the generation of a Zeta potential and EOF
When a voltage is applied across the capillary, the cations in the diffused layer will moved
towards the cathode pulling with them the bulk solution in the capillary. This is called electro-
osmotic flow (EOF). The charge on the capillary wall is highly dependant on the pH, which
means the EOF, is also dependant on the pH, therefore, the higher the pH the greater the EOF.
A benefit of EOF is its characteristic flat-flow profile, which results in sharp peaks with good
resolution (Figure 15). External pump systems used in HPLC result in laminar-flow profiles
with rounded broad peaks.
35
Figure 15. Flow profiles of EOF and laminar flow
y most of
e modern instruments have a cooling facility to overcome the problem of heating.
modes include capillary zone electrophoresis and
icellar electrokinetic chromatography.
A great advantage of EOF is that it causes migration of not only the cations, but because of
the flow of the buffer, the anions and the neutral molecules will also be moved towards the
cathode and the detector. The other factor affecting the mobility is the viscosity of the buffer.
With the passage of an electrical current through an electrolyte buffer, heat is generated and
this causes an elevation of temperature within the capillary. This heat causes a change in the
viscosity of the buffer. Control of the temperature is very important as a 1oC change in
temperature can result in a 3% change in viscosity, and thus a 3% change in mobility.
Temperature increase depends on the voltage applied to the system. Thus, by lowering the
applied voltage, a drop in temperature can be achieved, but the theoretical equation for
resolution and efficiency advocate the use of as high an electrical field as possible. A
reduction in the diameter of the capillary will cause a dramatic decrease in current, this will
cause a decrease in power generated, as well as in temperature. However, a reduction in the
diameter of the capillary will affect the sensitivity (Heiger, 1992). A decrease in the ionic
strength of the buffer can also be used to decrease the electrical current. Luckily toda
th
Capillary electrophoresis comprises of a number of different operation modes that have
different separation characters. Theses
m
36
2.9.2.2 Capillary zone electrophoresis
Capillary zone electrophoresis (CZE) is probably the most commonly used CE method
because of its simplicity and versatility. Samples are injected onto a narrow bore fused silica
capillary (25 - 75 mm ID) and separations of the analytes are dependant upon the different
migration times of the ionic species in the sample. The ends of the capillary are placed in
separated buffer reservoirs and the capillary is filled with the buffer. Electrodes are positioned
in the two reservoirs and connected to a high voltage power supply. Most of the time the
samples are loaded onto the capillary at the anode and the detection of the analytes take place
at the opposite end of the capillary. The effective mobility and therefore the separation of the
analytes are dictated by their charge to mass ratio at a specific pH. However, the migration
velocity of the different ions is dependant on the sum of the EOF, which is the bulk flow of
the liquid in the capillary, and their respective electrophoretic mobilities. Cations with the
greatest charge to mass ratio migrate first, followed by cations with a smaller ratio, then
neutral molecules, followed by anions with a smaller charge to mass ratio and lastly anions
with the largest charge to mass ratio.
2.9.2.3 Micellar electrokinetic chromatography
Micellar electrokinetic chromatography is normally used to resolve both charged and neutral
molecules in a single run. The basic principal of MEKC is the use of surfactants that are
incorporated into the buffer at concentrations above the critical micelle concentration. The
most commonly used surfactant is sodium dodecyl sulphate, an anionic salt. Charged micelles
migrate either with or against the EOF, depending on their charge. In the case of anionic
surfactants, the negatively charged head groups tend to orientate themselves on the outer
surface of the micelle, with the hydrophobic tail groups orientating themselves towards the
centre of the micelle. These anionic micelles are attracted to the anode, but because of the
EOF moving towards the cathode, they slowly move towards the cathode. If the analytes are
charged, they will migrate according to their electrophoretic mobilities, but neutral analytes
will migrate with the EOF and the micelles. The more hydrophilic the neutral analyte is, the
less time it will spend inside the micelle, and the quicker it will migrate with the EOF towards
the cathode. On the other hand the more hydrophobic the neural analyte is, the more time it
37
will spend in the micelle. Extremely hydrophobic compounds will remain in the micelle and
will elute with the micelle.
2.9.2.4 Capillary electrophoresis instrument
The modern capillary electrophoresis instrument is a very simple design (Figure16).
Figure 16. Schematic representation of the arrangement of the main components of
a capillary electrophoresis instrument
The two ends of a fused silica capillary column are placed into two buffer reservoirs, each
containing an electrode connected to a power supply. Samples are injected onto the capillary
by putting the one end of the capillary into the sample solution and applying either an
electrical potential or external pressure for a few seconds to move the sample into the
capillary. After this the capillary end is put back into the buffer reservoir and an electrical
potential is applied for the duration of the analyses. Detection is normally achieved through a
small window, burned into the capillary, near the opposite end from where the injection took
place. The most frequently used detector is a UV absorbance detector that is connected to a
data processor.
38
2.9.2.5 Capillaries The ideal properties for a capillary would include being chemically, physically and
electrically inert, as well as UV-Visible transparent, flexible, robust and inexpensive. The
capillaries, which are normally used, are made from fused silica with an external cover of
polyimide to give them mechanical strength, as bare fused silica is extremely fragile. A small
portion of this coating is usually removed to form a window for detection purposes. This
window is aligned in the optical centre of the detector. Capillaries are typically 25-100 cm
long with an internal diameter of 50-75 µm. On standard commercial CE instruments, the
capillary is held in a housing device to facilitate ease of capillary insertion into the instrument
and to help with the temperature control of the capillary. Coating different substances onto the
inner wall can also chemically modify the inner surface of the capillary. These coatings are
used for a variety of purposes, such as to reduce sample absorption or to change the ionic
charge of the capillary wall.
2.9.2.6 Electrolyte system
The electrolyte used for a specific analysis is of critical importance as its composition
determines the migration behaviour of the analytes. Because of the strong dependence of EOF
and electrophoretic mobilities on the electrolyte system, careful consideration of several
factors is necessary prior to the selection of a specific electrolyte system. The selected buffer
should ideally possess the following properties:
• Sufficient buffering capacity for the pH working rang; • Low absorbance at the detection wavelength;
• Temperature fluctuation must not effect its composition.
A wide range of electrolyte systems has been used to get the required separation, the majority
of these being aqueous buffers. In order to perform the electrophoretic separation, the analytes
must be soluble in the buffer. The ionic strength and the pH of the buffer also play an
important role in the selection of the electrolytic system. Fatty acids with chain lengths of
more than 14 carbons are insoluble in aqueous buffers. Thus, for the analyses of fatty acids
the buffers must consist of at least some sort of organic compound.
39
Care must also be taken that the buffer levels in both the anodic and cathodic reservoirs
remain at the same level. If the height of the buffer in the two reservoirs is not equal, a
pressure difference will result and siphoning will occur, effecting migration times. Great care
should be taken to restrict buffer depletion caused by electrolyses and ion migration. Buffer
depletion results in pH changes, which is probably the single parameter with the greatest
influence on separation. The pH of the buffer determines the electric charge of the analytes
and their electrophoretic mobility, as well as the charge on the silanol groups at the capillary
wall, and consequently, the EOF (Heiger, 1992).
2.9.2.7 Sample introduction The two most often-used injection methods are those of electrokinetic and hydrodynamic
injection. Using the electrokinetic injection, the electrode and capillary are inserted into the
sample vial and a voltage is applied for a few seconds. Field strengths about 5 times lower
than that used for separation, is usually used. This method tends to have greater precision than
that of the hydrodynamic technique, but it is not as reproducible. Hydrodynamic injection can
be accomplished by one of two methods. Pressure can be applied at the injection end or with
the application of a vacuum at the exit end of the capillary. The main advantage of
hydrodynamic injection is that there is no discrimination between different sample species
upon injection.
2.9.2.8 Detectors
UV absorbance detectors are most frequently used.
40
CHAPTER 3
MATERIALS AND METHODS FOR GLC
3.1 Introduction
The complete process of fatty acid analyses by GLC consists of the extraction and
transmethylation of the lipids, the injection, separation, identification and quantification of the
FAME’s. To achieve the required accuracy and precision, each process has to be optimised. In
this Section different extraction and methylation procedures are evaluated and the procedures
that give the best recovery of the extracted and transmethylated fatty acids will respectively
be used to optimise the separation and final analyses of the samples. Aliquots of the samples
will also be separated into their cis and trans mono-unsaturated fatty acid isomer fractions on
Ag-TLC. These fractions will also be analysed by GLC to evaluate to what extent the cis and
trans isomers overlapped when not pre-separated by Ag-TLC.
3.2 Sampling and sample handling Eighteen different brands of hard block margarines, including brands widely used by local
consumers, were bought from the local supermarket and stored at 4oC until they could be
analysed. On the day of the analyses about 100 g of each product was heated to 25oC in an
oven for 30 minutes. Each sample was thoroughly homogenised with a hand-held electric
mixer. From these homogenised samples, aliquots were taken for the analyses.
3.3 Chemicals and gases Analytical reagent grade chloroform, methanol and hexane (Merck Darmstadt, Germany)
were re-distilled in an all glass system. All glassware was rinsed with re-distilled methanol
and air-dried. All chemicals were analytical grade (Merck Darmstadt, Germany and Sigma
Chemical CO. St. Louis, MO 63178 USA). The fatty acid standards were certified to be
> 99% pure and purchased from Nu Chek Prep. (Nu- Chek- Prep, INC. Elysian, Minnesota,
41
USA). All gases employed (N , H2 2, and medical air) were of 99.99% purity (Liquid Air,
South Africa).
3.4 Evaluation of different lipid extraction solutions
The three extraction solutions chosen to evaluate the extraction of triacylglycerols were: a)
chloroform/methanol (C:M) (2:1), b) chloroform/methanol (C:M) (1:1), and c) n-hexane. For
the evaluation of the recovery of triacyglycerols, a test triacylglycerol sample was prepared by
dissolving 102.05 mg of a > 99% pure glyceryl triheptadecanoate acid standard, (Sigma
Chemical CO. St. Louis, MO 63178 USA) in 100 ml n-heptane. Glyceryl triheptadecanoate
acid standard was used because a trans triacylglycerol standard was not available for the
evaluation of the extraction procedure. The triacylglycerol standard that was used consisted of
a glycerol with three heptadecanoic acid molecules (17:0) attached to it. Because of the
structure resemblance between trans unsaturated fatty acids and saturated fatty acids, and the
differences in structure between cis and trans unsaturated fatty acids, it was decided to use a
triacylglycerol standard, formed from three saturated fatty acid molecules. The geometrical
structure of the test saturated fatty acids resembled those of trans fatty acids.
Nine aliquots of the test sample were precisely pipetted into 50 ml extraction tubes. Lipid
extracts were prepared by homogenising three of the test samples in 40 ml of C:M (2:1 v/v)
containing 0.01% BHT, another three test samples in 40 ml C:M (1:1 v/v) containing 0.01%
BHT, and the last three in 40 ml n-hexane with 0.01% BHT, by using a polytron (Kinematica,
type PT 10-35, Switzerland). After homogenising, all the homogenates were filtered with
sintered glass funnels. The funnels were washed with 5 ml of the different extraction solutions
and the filtrates made up to 50 ml in volumetric flasks with their respective solutions.
Quantitative aliquots, to give precisely 19.5 µg heptadecanoic acid (17:0), were taken from all
nine volumetric flasks and FAMEs were prepared using an in-house transmethylation method
based on the procedure described by Christie (1990). After cooling, 2 ml hexane and 1 ml
water was added to all the samples. The solutions were thoroughly mixed on a Vortex mixer
and the top hexane layers containing the FAMEs were transferred to glass tubes. The
extraction procedure was repeated three times and the respective hexane phases pooled.
To each of the pooled test samples, 20.0 µg of trans-9, octadecenoic acid (trans-9, 18:1)
methyl ester reference standard (Nu- Chek- Prep, INC. Elysian, Minnesota, USA) was added
42
as an internal standard, and the solutions were evaporated to dryness under a stream of
nitrogen gas in a water bath at 40oC. The residues were re-dissolved in 50 microliter carbon
disulfide (CS ) and one microliter was subjected to GLC analyses. 2
To verify these, by washing the samples three times with hexane, all the FAMEs were
recovered from the samples; a further 2 ml of hexane was added to each of the nine sample
extraction tubes and extracted again. All the hexane layers were pooled into a separate tube
and evaporated to dryness. This pooled sample was analysed, with the other nine samples.
The FAMEs in all the samples were identified by GLC as described by Ball et al. (1993)
using two Varian Model 3300 GLCs fitted with BPX-70 capillary columns. The one was
fitted with a 30 m BPX-70 fused silica capillary column with an internal diameter 0.32 mm
coated with 70% Cyanopropyl polysilphenylene-siloxane to a thickness of 0.25 µm (SGE
International Pty Ltd, Australia). The other was fitted with a 120 m BPX 70 fused silica
capillary column with an internal diameter of 0.25 mm also coated with 70% Cyanopropyl
polysilphenylene-siloxane to a thickness of 0.25 µm (SGE International Pty Ltd, Australia).
Both instruments were equipped with flame ionisation detectors. The analyses were done with
isothermal column temperatures of 180o -1C and column gas flow rate of 30 cm sec . Gas flow
rates were: hydrogen, 25 ml/min and air 250 ml/min. The injector temperatures were 240oC
and detector temperatures 280oC. One microliter samples were injected manually at a split
ratio of 1:80 (Ball et al., 1993).
3.5 Evaluation of lipid transmethylation procedures For the evaluation of the best transmethylation method to be used for the preparation of
FAMEs of triacylglycerols, a test sample was prepared by dissolving a > 99% pure glyceryl
triheptadecanoate acid standard (Sigma Chemical CO. St. Louis, MO 63178 USA) in n-
heptane. This test sample was extracted with the extraction solvent that gave the best recovery
of the FAMEs as determined under Section 3.4. Quantitative aliquots (28.5 µg) from the
extracted test sample were pipetted into six methylation tubes. To three of these, 2 ml of 5%
concentrated sulphuric acid in re-distilled methanol (v/v) was added, then sealed with Teflon-
lined caps and heated in a metal block for two hours at 70oC. Thereafter they were cooled to
room temperature (Christie, 1990). To the other three sample tubes, 5 ml of 0.5 M sodium
43
methoxide in anhydrous methanol (Aldrich Chemical CO. INC. Milwaukee, WI 53201 USA.)
was added and sealed with Teflon-lined caps. These tubes were heated in a 40oC water bath
for 5 minutes, and after cooling a drop of concentrated glacial acetic acid were added to each
tube to neutralise the reaction (Richardson et al., 1997). One millilitre of water and 2 ml of
hexane were added before the tubes were vortexed for 30 seconds. Thereafter the upper
hexane layers, containing the methyl esters, were transferred to extracting tubes. The
extraction procedure was repeated three times and the respective hexane phases pooled. To all
six tubes, 20.0 µg of trans-9, octadecenoic acid (trans-9, 18:1) methyl ester reference
standard (Nu- Chek- Prep, INC. Elysian, Minnesota, USA), as an internal standard, was added
and mixed well before the extracts were evaporated to dryness under a stream of nitrogen gas
in a 40oC water bath. These samples were also analysed as described under Section 3.4.
To evaluate the effect of the different sample concentrations to a constant transmethylation
solution volume, three triplicate triacylglycerol test samples with concentrations of
19.5 µg/100µl, 28.5 µg/100µl and 59 µg/100µl were transmethylated with the two
transmethylation solutions under investigation. After transmethylation, the samples were
extracted as described earlier. To all nine tubes, 40 µg of trans-9, octadecenoic acid methyl
ester reference standard solution was added and evaporated to dryness before subjected to
GLC analyses.
3.6 Sample preparation
After the evaluation of the different extraction and transmethylation solutions, the methods
giving the best recoveries were used for the preparation of the margarine samples for GLC
analyses. The fatty acid constituents of the margarines were identified and quantified by
accurately weighing 300 mg aliquots of the homogenised samples into extraction tubes, and a
known concentration of glyceryl triheptadecanoate acid (17:0), as an internal standard, was
added to the samples. Lipid extracts were prepared by homogenising the samples with a
polytron (Kinematica, type PT 10-35, Switzerland) in 40 ml of the extraction solution that
gave the best recoveries of the three solutions evaluated. After homogenising, the
homogenates were filtered with sintered glass funnels. The funnels were washed with 5 ml of
the extraction solution, and the filtrates made up to 50 ml with the chosen extraction solution
in 50 ml volumetric flasks. Two millilitres of the transmethylation solution, which gave the
44
best recoveries of the two methods evaluated, were added to 500 µl of the different sample
extracts, then sealed with Teflon-lined caps and transmethylated as described by Christie
(1990). After cooling, 1 ml water and 2 ml hexane were added, and the samples thoroughly
mixed on a Vortex mixer. The samples were extracted once only with hexane, because of the
internal standard that was added to the sample before the extraction solution. The assumption
can be made that if there is any loss of the sample, the same will happen to the internal
standard. The top hexane layers were evaporated to dryness and re-dissolved in CS2 before
GLC analyses.
3.7 Evaluation of the two BPX-70 GLC columns
Two columns were evaluated. The one, a 30-m BPX-70 capillary column, is normally used
for routine fatty acid analyses, while the other one is a 120-m BPX-70 capillary column. For
the separation of the different cis and trans fatty acid isomers, most of the authors
recommended a long column (Precht et al., 1996; Aro et al., 1998; Ball et al., 1993). A
pooled margarine FAME sample was prepared for the evaluation of the two capillary
columns. Two Varian Model 3300 GLCs fitted with these two columns were used for the
identification of the FAMEs in the pooled sample. The one was fitted with a 30-m BPX-70
fused silica capillary column with an internal diameter of 0.32 mm and coated with 70%
Cyanopropyl polysilphenylene-siloxane to a thickness of 0.25 µm (SGE International Pty Ltd,
Australia). The other was fitted with a 120-m BPX 70 fused silica capillary column with an
internal diameter of 0.25 mm, also coated with 70% Cyanopropyl polysilphenylene-siloxane
to a thickness of 0.25 µm (SGE International Pty Ltd, Australia). Both instruments were
equipped with flame ionisation detectors. The evaluation of the two columns was done with
isothermal column temperatures of 180o -1C and a hydrogen column flow rate of 30 cm sec .
Gas flow rates were: hydrogen, 25 ml/min and air 250 ml/min. The injector temperatures were
240o oC and detector temperatures 280 C (Ball et al., 1993). One microlitre samples were
injected manually at a split ratio of 1:80. A computer fitted with a Delta integration program
(Dataworx Pty Ltd, 17/1 Goodwin St. Kangaroo Point, Brisbane, Australia.) controlled the
GLC systems and did the integration of the peaks.
45
3.8 Identification of the different standard isomers
After the evaluation and selection of the column that gives the best separation of the different
fatty acids in the pooled sample, the identification and elution order of the different cis and
trans fatty acid isomers were done by the different retention times of cis and trans FAME
standard isomers (Nu- Chek- Prep, INC. Elysian, Minnesota, USA). To further assist with the
identification, Equivalent Chain-Length (ECL) values for cis and trans FAMEs from SGE
Analytical Science (www.sge.com) were also used. The FAME standards were evaporated to
dryness under a stream of nitrogen in a 40oC water bath. The residues were then re-dissolved
in CS2 and analysed (Ball et al., 1993). Different column gas flow rates between 28 and
38 cm sec-1 and column temperatures between 151o oC and 191 C were used to evaluate these
factors on the elution order of the different standard isomers, as well as their effect on the
separation power of the column.
After the selection of the best column length to use for the analyses of the margarine samples
and the identification and elution order of the different standard cis and trans isomers, the
pooled margarine FAME sample was injected on the chosen column. Different column
temperatures between 151o oC and 197 C and different hydrogen column flow rates between 28
and 38 cm sec-1 were used to evaluate these effects on the separation of real margarine
FAMEs, which normally has many more different fatty acids.
Aliquots of the same pooled sample were also subjected to Ag-TLC fractionation and the cis
and trans mono-unsaturated FAME fractions were also subjected to GLC analyses.
3.9 Evaluation of silver ion thin layer chromatography
Silver ion thin layer chromatography is use to separate the FAMEs of a sample into its
saturated, cis mono-unsaturated, trans mono-unsaturated and polyunsaturated fatty acid
fractions (Precht et al., 1996).
Glass (20 x 20 cm) thin layer chromatography (TLC) plates (Merck, Darmstadt, Germany)
coated with 0.25 mm Silica Gel 60 were dipped into a 10% (w/v) silver nitrate aqueous
solution for 20 minutes, air dried and stored in a dark room. Before use, the plates were
46
activated at 120oC for 30 minutes and used within one hour after cooling (Precht et al., 1996).
Leaving a space of about 2 cm at both edges, the pooled FAME sample dissolved in n-heptane
was applied to the plate in a narrow band. Two methyl ester standards, a cis and a trans
methyl ester were also applied in the same manner. After developing with n-heptane/diethyl
ether (90:10) in a TLC chamber lined with filter paper, the plate was air dried and the
fractions were visualised by lightly spraying the plates with a 0.2% solution of 2,7-
dichlorofluorescein in iso-propanol and marked under ultra violet (UV) light. The cis and
trans mono-unsaturated fatty acid fractions were identified by the co-migration of the
standards, scraped off separately and eluted three times with diethyl ether. The eluents were
pooled and evaporated to dryness under a stream of nitrogen and re-dissolved in CS2 before
injecting this into the GLC.
After the optimisation of the carrier gas flow rate, the column temperature and the evaluation
of the Ag-TLC results the samples were analysed.
47
CHAPTER 4
GLC RESULTS AND DISCUSSION
4.1 Evaluation of the different extraction solvents
Extraction is the first important step in the preparation of fatty acid methyl esters for the
identification and quantification of fatty acids. The results of the different extraction solutions
will be used to select an extraction solution that would give the best recovery of
triacylglycerols.
The chromatograms from the GLC fitted with a 30 m BPX-70 column demonstrate the results
of the triacylglycerol test sample, extracted with chloroform/methanol (C:M) (2:1) (Figure
17.1), chloroform/methanol (C:M) (1:1) (Figure 17.2) and n-hexane (Figure 17.3).
minutes
mill
iVol
ts
2.5 5.0 7.5 10.0 12.5 15.0 17.5
0
1000
2000
1
2 3
minutes
mill
iVol
ts
2.5 5.0 7.5 10.0 12.5 15.0 17.5
0
1000
2000
1
2 3
Figure 17.1. Chromatogram of test sample and internal standard using chloroform:
methanol (2:1) as the extraction solution and a 30 m BPX-70 column at 180oC Peak identification: 1= CS2; 2= 17:0; 3= trans-9, 18:1
48
minutes
mill
iVol
ts
2.5 5.0 7.5 10.0 12.5 15.0 17.5
0
1000
2000
1
2 3
minutes
mill
iVol
ts
2.5 5.0 7.5 10.0 12.5 15.0 17.5
0
1000
2000
1
2 3
Figure 17.2. Chromatogram of the test sample and internal standard using chloroform:
methanol (1:1) as the extraction solution and a 30 m BPX-70 column at 180oC Peak identification: 1= CS2; 2= 17:0; 3= trans-9, 18:1
minutes
mill
iVol
ts
2.5 5.0 7.5 10.0 12.5 15.0 17.5
0
1000
2000
1
2 3
minutes
mill
iVol
ts
2.5 5.0 7.5 10.0 12.5 15.0 17.5
0
1000
2000
1
2 3
Figure 17.3. Chromatogram of the test sample and internal standard using hexane as the
extraction solution and a 30 m BPX-70 column at 180oC Peak identification: 1= CS2; 2= 17:0; 3= trans-9, 18:1
49
The peak heights of the test samples and the internal standards in all the chromatograms were
less than 2000 millivolts and, therefore, well within the maximum analogue output of 10 000
millivolts of the GLCs. All measurements were also done in the linear range of the detector.
The maximum range of the FID detectors is 107 nano Ampere (nA), as specified by the
manufacturers. For the detection of the samples the FID range was set to 105 nA, well within
the accepted linear range of the FID.
The two peaks representing the test sample and the internal standard using the three extraction
solvents are well resolved with good baseline separation. The retention times of the two
FAMEs used for the evaluation (test sample and the internal reference standard,) using the
different extraction solvents were the same. The peaks were sharp with no tailing, indicating
that the column was not overloaded.
The chromatograms of the three extraction methods using a 120 m BPX-70 column (Figures
17.4-17.6) show that the length of the column is an important parameter. With a 30 m BPX-70
column, the retention time of the trans-9, 18:1 internal standard is about 10.5 minutes, while
using the same GLC conditions, but with a 120 m BPX-70 column, the retention time
increases to just under 50 minutes. Thus, an increase in the column length increases the
retention times.
minutes
milli
Volts
0.0 20.0 50.00
1000
2000
10.0
12
3
30.0 40.0minutes
milli
Volts
0.0 20.0 50.00
1000
2000
10.0
12
3
30.0 40.0
Figure 17.4. Chromatogram of the test sample and internal standard using chloroform:
methanol (2:1) as the extraction solution and a 120 m BPX-70 column at 180oC Peak identification: 1= CS2; 2= 17:0; 3= trans-9, 18:1
50
minutes
milli
Volts
0.0 20.0 50.00
1000
2000
10.0
12
3
30.0 40.0 60.0minutes
milli
Volts
0.0 20.0 50.00
1000
2000
10.0
12
3
30.0 40.0 60.0
Figure 17.5. Chromatogram of the test sample and internal standard using chloroform:
methanol (1:1) as the extraction solution and a 120 m BPX-70 column at 180oC Peak identification: 1= CS2; 2= 17:0; 3= trans-9, 18:1
minutes
milli
Vol
ts
0.0 20.0 50.00
1000
2000
10.0
12
3
30.0 40.0 60.0minutes
milli
Vol
ts
0.0 20.0 50.00
1000
2000
10.0
12
3
30.0 40.0 60.0
Figure 17.6. Chromatogram of the test sample and internal standard using hexane as the
extraction solution and a 120 m BPX-70 column at 180oC Peak identification: 1= CS2; 2= 17:0; 3= trans-9, 18:1
51
The large difference in retention times between the 30 m and the 120 m columns is normal.
The two columns used in this study were capillary columns with a small internal diameter
(ID). The ID of the 30 m column was 0.32 mm and that of the 120 m column, 0.25 mm. The
separation power of GLC capillary columns is given by the total chromatographic plate count.
The average plate count for capillary columns of 0.32 mm ID, is about 2 500 plates/meter and
for 0.25 mm ID, about 3 300 plates/meter. Thus, the longer and the smaller the internal
diameter of the column, the more plates it will have and the greater the interaction of the
sample components with the stationary phase. The greater the interaction, the slower the
sample compounds will migrate towards the detector. Using a longer column with a small ID
will retain the different compounds longer and thus extend the elution ranges of the
compounds in the samples. In this way, the elution times or retention times of the different
compounds are pulled apart, and a greater separation space will be available to the
components in the sample eluting close together. It is well known that the different cis and
trans octadecenoic acid (18:1) isomers within oils eluted in a narrow time range (Kramer et
al., 2004). This time range can be increased by increasing the column length as is well
demonstrated by the differences in the retention times between the two peaks of the test
standard and the internal standard using the two lengths. With a 30 m column the time
difference between the two peaks is 2 minutes, while with the 120 m column, it is nearly 15
minutes with a column temperature of 180oC and a hydrogen carrier gas flow rate of 30 cm
sec-1.
The sample recovery results using the three extraction methods and a 30 m BPX-70 column
are summarised in Table 2, while those using a 120 m BPX-70 column are summarised in
Table 3. Graph 1 graphically illustrates the results.
52
Table 2. Recovery results of the test sample, as determined by the area counts of triplicate extractions using chloroform/methanol (2:1), chloroform/methanol (1:1) and hexane as the extraction solutions, injected into a 30 m BPX-70 column
Solvent No 17:0 18:1 Avg. STD. %CV True Measured Avg. % Accuracy
Rec. value value A 3084084 3297282 19.5 18.7
C:M(2:1) B 5417629 5590713 95.8 2.0 2.1 19.5 19.4 19.2 98.5C 3782215 3894780 19.5 19.4
A 2483916 2943096 19.5 16.9C:M(1:1) B 2721154 3302212 82.7 1.6 1.9 19.5 16.5 16.5 84.6
C 3006318 3698870 19.5 16.3
A 3122132 3435303 19.5 18.2Hexane B 2593123 2905583 90.4 1.0 1.1 19.5 17.8 18.1 92.8
C 4143872 4552043 19.5 18.2
Table 3. Recovery results of the test sample, as determined by the area counts of triplicate extractions using chloroform/methanol (2:1), chloroform/methanol (1:1) and hexane as the extraction solutions, injected into a 120 m BPX-70 column
Solvent No 17:0 18:1 Avg. STD. %CV True Measured Avg. % Accuracy
Rec. value value A 13081019 13786297 19.5 19.0
C:M(2:1) B 15940621 16876832 95.4 1.3 1.4 19.5 18.9 19.1 97.9C 21377090 22061821 19.5 19.4
A 16636799 19117020 19.5 17.4C:M(1:1) B 18745484 21225704 87.9 0.7 0.8 19.5 17.7 17.6 90.2
C 22846331 25867776 19.5 17.7
A 18831374 20657706 19.5 18.2Hexane B 37004800 39031132 93.3 1.9 2.0 19.5 19.0 18.7 95.9
C 28727594 30614273 19.5 18.8
53
80.00
82.00
84.00
86.00
88.00
90.00
92.00
94.00
96.00
98.00
100.00
30m 120m 30m 120m 30m 120m
C:M(2:1) C:M(1:1) Hexane
% R
ecov
ery
Graph 1. The percentage recovery of the test sample using a 30 m and a 120 m BPX-70
capillary column and chloroform:methanol (2:1), chloroform:methanol (1:1) and
hexane as the extraction solutions
With a 30 m column, the average recovery using C:M (2:1) was 95.8 ± 2.0%, C:M (1:1) was
82.7 ± 1.6%, and with hexane it was 90.4 ± 1.0%. With the 120 m column, the average
recovery was 95.4 ± 1.3%, with C:M (2:1), 87.9 ± 0.7%, with C:M (1:1) and with hexane it
was 93.3 ± 1.9%. The coefficient of variants of all three methods using both columns was
below 5%. Of the three extraction methods, the accuracy using the C:M (2:1) extraction
solution, on both columns, was the nearest to 100 %. The differences in sample recovery as
illustrated in Graph 1, between a 30 m and 120 m column, using the same extraction solution,
cannot be explained. Graph 1 also illustrates that the C:M (2:1) extraction solution gave the
best recovery, despite the column length, while C:M (1:1) gave the lowest average recovery
on both columns. This can be because of the chloroform/methanol ratio difference between
the two extraction solutions. For the identification and quantification of individual fatty acids
without isomers, a 30 m column gives good results in a short analytical time. The retention
time of the trans-9, 18:1 isomer, used as an internal reference standard was less than 11
minutes using a 30 m column, while it increased to 50 minutes with a 120 m column. This
longer analytical time will have a negative effect on the number of samples that can be
analysed during a working day. On the other hand, a longer column with more theoretical
plates will retain the different compounds longer and thus extend the elution ranges of the
54
compounds in the samples. In this way, the retention times of the different compounds are
pulled apart and a greater separation space is available for the fatty acid isomers eluting
closely together, thereby contributing to less overlapping of the peaks.
The chromatogram in Figure 17.7 shows the results of the final pooled hexane phases. The re-
extracted residues yielded no detectable peaks proving that all the FAMEs were extracted
during the first three hexane extractions. By doing this confirms that washing the samples
with hexane three times, extracts all the FAMEs. However, this does not guarantee that all the
fatty acids are transmethylated.
minutes
milli
Volts
0.0 20.0 50.00
1000
2000
10.0
1
30.0 40.0 60.0minutes
milli
Volts
0.0 20.0 50.00
1000
2000
10.0
1
30.0 40.0 60.0
Figure 17.7. The chromatogram of the final pooled hexane extractions to verify that
washing the samples three times with hexane recovered all FAME in the nine test
samples
Peak identification: 1= Solvent peak
Based on the improved test sample recoveries using C:M (2:1, v/v) with both column lengths,
it was decided to use C:M (2:1, v/v) as the preferred solution for the extraction of the
triacylglycerols in the margarine samples.
55
4.2 Evaluation of the two transmethylation solvents
The evaluation of the different extraction solutions showed that C:M (2:1) gave the best
recovery of the test sample and for the evaluation of the two transmethylation solvents the test
samples were extracted with C:M (2:1).
The chromatograms (Figures 18.1-18.4) that were generated when using 30 m and 120 m
BPX-70 capillary columns with the two transmethylation reagents, showed that the peaks of
the test sample and the internal FAME reference standard were well resolved and sharp. The
retention times of the two peaks using a 30 m column were the same for the two
transmethylation reagents. The same observation was made using a 120 m column, except that
the retention times were obviously longer.
minutes
5.0 7.5 10.0 12.5 15.0
12 3
112 3
12 3
1
2000
1000
0
mill
iVol
ts
2.5
2 3
17.5
minutes
5.0 7.5 10.0 12.5 15.0
12 3
112 3
12 3
1
2000
1000
0
mill
iVol
ts
2.5
2 3
17.5
Figure 18.1. Chromatogram of the test sample using 5% concentrated sulphuric acid in
methanol as the transmethylation reagent with a 30 m BPX-70 column at 180oC
Peak identification: 1= CS2; 2= 17:0; 3= trans-9, 18:1
56
minutes
5.0 7.5 10.0 12.5 15.0
12 3
112 3
12 3
1
2000
1000
0
mill
iVol
ts
2.5
2 3
17.5
minutes
5.0 7.5 10.0 12.5 15.0
12 3
112 3
12 3
1
2000
1000
0
mill
iVol
ts
2.5
2 3
17.5
Figure 18.2. Chromatogram of the test sample using 0.5 M methoxide as the
transmethylation reagent, with a 30 m BPX-70 column at 180oC Peak identification: 1= CS2; 2= 17:0; 3= trans-9, 18:1
1
2000
1000
0
mill
iVol
ts
2 3
minutes20.0 30.0 40.0 50.010.0
1
2000
1000
0
mill
iVol
ts
2 3
minutes20.0 30.0 40.0 50.010.0
Figure 18.3. Chromatogram of the test sample using 5% concentrated sulphuric acid in
methanol as the transmethylation reagent with a 120 m BPX-70 column at 180oC Peak identification: 1= CS2; 2=17:0; 3= trans-9, 18:1
57
1
2000
1000
0
mill
iVol
ts
2 3
minutes20.0 30.0 40.0 50.010.0
1
2000
1000
0
mill
iVol
ts
2 3
minutes20.0 30.0 40.0 50.010.0
Figure 18.4. Chromatogram of the test sample using 0.5 M methoxide as the
transmethylation reagent with a 120 m BPX-70 column at 180oC Peak identification: 1= CS2; 2= 17:0; 3= trans-9, 18:1
The sample recovery results using the two transmethylation methods and a 30 m BPX-70
column are summarised in Tables 4 and the results using a 120 m BPX-70 column are
summarised in Table 5.
Table 4. Recovery results as determined by the GLC area counts of triplicate extractions when using 5% sulphuric acid/methanol and 0.5 M sodium methoxide/methanol transmethylation solvents and a 30 m BPX-70 column
Method No 17:0 18:1 Avg. Std %CV True Measured Avg. % AccuracyRec. Value Value
A 5602292 4005386 28.5 28.0H2SO4 B 4476612 3284426 97.5 0.5 0.5 28.5 27.3 27.8 97.5
C 4833101 3436195 28.5 28.1
A 3500466 2702509 28.5 25.9NaOCH3 B 2618442 2003865 90.5 0.4 0.4 28.5 26.1 25.8 90.5
C 4020185 3167891 28.5 25.4
58
Table 5. Recovery results as determined by the GLC area counts of triplicate extractions when using 5% sulphuric acid/methanol and 0.5 M sodium methoxide/methanol transmethylation solvents and a 120 m BPX-70 column
Method No 17:0 18:1 Avg. Std %CV True Measured Avg. % Accuracy
Rec. Value ValueA 28118396 19811446 28.5 28.4
H2SO4 B 30165464 21713654 98.5 0.4 0.4 28.5 27.8 28.0 98.2C 32183814 23183919 28.5 27.8
A 24791918 19283408 28.5 25.7NaOCH3 B 23882741 18372506 90.7 0.2 0.2 28.5 26.0 25.8 90.5
C 28765312 22346517 28.5 25.7
With a 30 m column, the average recovery was 97.5 ± 0.5% and on a 120 m column it was
98.5 ± 0.4% using 5% concentrated sulphuric acid in methanol. With 0.5 M sodium
methoxide in methanol the recovery was 90.5 ± 0.4% and 90.7 ± 0.2% on a 30 m and a 120 m
column, respectively. These results showed that 5% concentrated sulphuric acid in methanol
gave a ± 7% better sample recovery than 0.5 M sodium methoxide in methanol for both the
column lengths. The % accuracy of the 5% sulphuric acid/methanol transmethylation reagent
varied between 97.5% and 98.2% using a 30 m and a 120 m column, respectively. With 0.5 M
sodium methoxide in methanol the accuracy was 90.5% for both columns. These results
illustrated that there is some sample lost when using 0.5 M sodium methoxide in methanol as
the transmethylation reagent, and that the same loss was noticed using both column lengths. It
was clearly shown that 5% concentrated sulphuric acid in methanol, as a transmethylation
solution, gave the least biased results and a slightly better sample recovery on both columns.
59
84.00
86.00
88.00
90.00
92.00
94.00
96.00
98.00
100.00
30m 120m 30m 120m
Sulphuric acid/Methanol Methoxide/Methanol
% R
ecov
ery
Graph 2. The percentage recovery of the test sample using 5% sulphuric acid/ methanol
and 0.5 M sodium methoxide/methanol transmethylation reagents after analysis with
two GLCs equipped with 30 m and 120 m BPX-70 columns
Graph 2 shows the recoveries using 5% concentrated sulphuric acid in redistilled methanol as
the transmethylation reagent compared to 0.5 M sodium methoxide/methanol . For the routine
analyses of FAMEs a 30 m column is preferred considering the shorter analytical time, though
it will have to be tested whether a short column will have enough separation power to identify
and quantify the different cis and trans fatty acid isomers in a partially hydrogenated oil
sample. These isomers normally elute before and after cis-9, 18:1. The same graph shows that
a 5% concentrated sulphuric acid in redistilled methanol, as the transmethylation reagent,
gives ± 8% better test sample recovery than 0.5 M sodium methoxide/methanol on both
columns.
60
61
Because of the better recovery results generated by the 120 m column when using the two
transmethylation reagents, only the 120 m BPX-70 column was used to evaluate the ratio
effect of different sample concentrations to fixed volumes of the two transmethylation
reagents.
The results in Table 6 show that with a sample containing 19.0 µg triacylglycerol, the
recovery was 99.7 ± 1.7% with 5% sulphuric acid/ methanol and 95.9 ± 2.3% with 0.5 M
sodium methoxide/ methanol. However, with a sample containing 28.5 µg triacylglycerol, the
recoveries dropped to 98.1 ± 1.8% and 94.1 ± 2.1%, respectively. With a higher sample
concentration, the effect was even more pronounced. With a sample containing 57.0 µg
triacylglycerol, the recovery was 98.5 ± 1.6% using 5% sulphuric acid/ methanol, and with
0.5 M sodium methoxide/ methanol the recovery dropped to 91.3 ± 1.1%.
These results demonstrated that the recovery of samples with concentrations ranging from
19.0 µg to 57 µg, transmethylated with 2 ml of 5% concentrated sulphuric acid/ methanol
reagent, is almost 100%. With 5 ml of 0.5 M sodium methoxide/ methanol solution, the
recovery dropped from 95.9 ± 2.3% to just more than 90% with increasing sample
concentrations. This demonstrated that the ratio between the sample triacylglycerols
concentration and the volume of the transmethylation reagent had a notable effect on the
recovery of the samples when using 0.5 M sodium methoxide/ methanol as the
transmethylation solution.
Method Conc. No 17:0 18:1 Rec. % Rec. Avg. STD %CV Avg. % AccuracyRec.
A 28118396 19811446 56.8 99.6H2SO4 57.0 B 36781234 25968367 56.7 99.4 98.5 1.6 1.7 56.2 98.6
C 67342156 48894532 55.1 96.7
A 8867279 12671829 28.0 98.2H2SO4 28.5 B 10908743 15342318 28.4 99.8 98.1 1.8 1.8 28.0 98.2
C 20132314 29345389 27.4 96.3
A 9566304 19887654 19.2 101.3H2SO4 19.0 B 15432314 33213156 18.6 97.8 99.7 1.7 1.8 18.9 99.5
C 12349548 25983210 19.0 100.1
A 24791918 19283408 51.4 90.2NaOCH3 57.0 B 17654321 13564309 52.1 91.3 91.3 1.1 1.2 52.1 91.4
C 32345821 24567892 52.7 92.4
A 6475466 9875356 26.2 92.0NaOCH3 28.5 B 13498760 19675438 27.4 96.3 94.1 2.1 2.3 26.8 94.0
C 29874523 44567234 26.8 94.1
A 6380346 14153521 18.0 94.9NaOCH3 19.0 B 10987694 23458712 18.7 98.6 95.9 2.3 2.4 18.2 95.8
C 21348125 47685432 17.9 94.2
Table 6. Recovery results of different test samples concentrations, using 5% sulphuric acid/ methanol and
0.5 M sodium methoxide/ methanol reagents as the two transmethylation reagents
62
8 6
8 8
9 0
9 2
9 4
9 6
9 8
1 0 0
1 0 2
1 9 .0 µ g
% R
ecov
ery
S u lp h u ric a c id /M e th a n o l M e th o x id e /M e th a n o l
2 8 .5 µ g 5 7 .0 µ g
Graph 3. The percentage recoveries of different samples, using 5% sulphuric acid/
methanol and 0.5 M sodium methoxide/ methanol transmethylation reagents
Graph 3 graphically illustrates the concentration effect. The effect is not so drastic using 5%
sulphuric acid/ methanol, but with 0.5 M sodium methoxide/ methanol there is a linear drop in
sample recoveries, with an increase in sample concentration. From these results, it is clear that
5% sulphuric acid/methanol transmethylation reagent would be the better choice when
working with samples containing an unknown fatty acid concentration. Although 0.5 M
sodium methoxide/methanol transmethylate triacylglycerols rapidly, a loss of FAMEs
occurred in samples with high triacylglycerols concentrations.
Based on the ± 8% better test sample recovery, as well as the nearly 100% recovery of
samples with different FAME concentrations, it was decided to use 5% concentrated sulphuric
acid in redistilled methanol (v/v) as the transmethylation reagent for the preparation of
FAMEs from the margarine samples.
63
4.3 Evaluation of the two columns
Although a 30 m BPX 70 column gave good separations with a short analysis time of the
normal occurring fatty acids, samples like partially hydrogenated oils with a large number of
different isomers, especially those occurring between 18:0 and 18:2, could not be separated
completely. This observation is supported by Kramer et al. (2004). The overlapping of the
different cis and trans 18:1 fatty acid isomers are illustrated in the chromatogram of the
pooled margarine sample in Figure19.1.
18:2
minutes
milli
Vol
ts
9.00 10.00 11.000
200
400
600
800
18:0
18:1
18:2
minutes
milli
Vol
ts
9.00 10.00 11.000
200
400
600
800
18:0
18:2
minutes
milli
Vol
ts
9.00 10.00 11.000
200
400
600
800
18:0
18:1
18:2
minutes
milli
Vol
ts
9.00
18:2
minutes
milli
Vol
ts
9.00 10.00 11.000
200
400
600
800
18:0
18:1
18:2
minutes
milli
Vol
ts
9.00 10.00 11.000
200
400
600
800
18:0
Figure 19.1. Part of a chromatogram showing the different 18:1 fatty acid isomers
(under the bracket) using a 30 m BPX 70 column at 180oC
From Figure 19.1 it is evident that some isomers are overlapping when using a 30 m BPX-70
column. Even the use of different column temperatures and column gas flow rates, did not
improve the separation. The analytical elution range of the different cis and trans 18:1
isomers are contracted into a too narrow retention time range for proper baseline separation.
The 120 m BPX-70 column with a smaller ID and about 220 000 theoretical plates, compared
to the 80 000 theoretical plates for a 30 m column (as specified by the manufacturer) provided
64
the required mechanism for extending the retention times of the different fatty acid isomers by
retaining the different compounds longer. In this way, the retention times of the different
isomers were pulled apart and a greater separation space became available to the different
isomers, allowing more isomers to be identified (see Figure 19.2). A disadvantage of using
longer columns was the longer analysis time per sample. However, this longer analytical time
increased the number of isomers that could be identified and quantified.
minutes
milli
Vol
ts
0
400
800
25.0 27.5 30.0 32.5
18:0
18:1
22.5
18:21000
800
minutes
milli
Vol
ts
0
400
800
25.0 27.5 30.0 32.5
18:0
18:1
22.5
18:21000
800
Figure 19.2. Part of a chromatogram showing the different 18:1 fatty acid isomers
(under the bracket) using a 120 m BPX 70 column at 180oC
Based on the inability of a 30 m BPX-70 column to separate most of the different cis and
trans 18:1 isomers, it was decided only to use a 120 m BPX-70 capillary column for the
identification and quantification of the different normal occurring fatty acids, as well as the
different cis and trans 18:1 isomers in the margarine samples.
65
4.4 Identification of the standard isomers
The GLC chromatograms (Figures 20.1-20.4) of the standard cis and trans FAME mixture,
using a 120 m BPX-70 capillary column, show the elution order to be as follows: trans-6,
trans-9 and trans-11 followed by cis-6, cis-9 and cis-11. The use of different column
temperatures between 151o oC and 191 C and different hydrogen column gas flow velocities
has no effect on the elution order of the different isomers in the standard mixture. This was
confirmed in a technical article by SGE on the analyses of 18:1 positional isomers using a 120
m BPX-70 capillary column (www.sge.com).
oC 1 2 3 4 5 6151
milli
Vol
ts
85.0
800
0
65.0 70.0 75.0 80.0
600
400
200
minutes
oC 1 2 3 4 5 6151
milli
Vol
ts
85.0
800
0
65.0 70.0 75.0 80.0
600
400
200
minutes
Figure 20.1. Part of the chromatogram showing the separation of six standard 18:1
isomers analysed at a column temperature of 151oC on a 120 m BPX-70 capillary
column
Peak identification: 1= t6, 2= t9, 3= t11, 4=c6, 5=c9, 6=c11
66
1 2 3 4 5 6171oC
milli
Volts
42.5
800
0
32.5 35.0 37.5 40.0
600
400
200
minutes
1 2 3 4 5 6
milli
Volts
42.5
800
0
32.5 35.0 37.5 40.0
600
400
200
minutes
1 2 3 4 5 6171oC
milli
Volts
42.5
800
0
32.5 35.0 37.5 40.0
600
400
200
minutes
1 2 3 4 5 6
milli
Volts
42.5
800
0
32.5 35.0 37.5 40.0
600
400
200
minutes
Figure 20.2 Part of the chromatogram showing the separation of six standard 18:1
isomers analysed at a column temperature of 171oC on a 120 m BPX-70 capillary
column
Peak identification: 1= t6, 2= t9, 3= t11, 4=c6, 5=c9, 6=c11
milli
Vol
ts
30.000
500
0
minutes
00
181oC
25.000
1 2 3 4 5 6
00
22.50 27.500
1000
1500
2000181 C
milli
Vol
ts
30.000
500
0
minutes
00
181oC
25.000
1 2 3 4 5 6
00
22.50 27.500
1000
1500
2000181 C
Figure 20.3. Part of the chromatogram showing the separation of six standard 18:1
isomers analysed at a column temperature of 181oC on a 120 m BPX-70 capillary
column
Peak identification: 1= t6, 2= t9, 3= t11, 4=c6, 5=c9, 6=c11
67
020.00
milli
Volts
22.50
500
1500
2500 o123 4 5 6
minutes
020.00
milli
Volts
22.50
500
1500
2500 191oC123 4 5 6
minutes
020.00
milli
Volts
22.50
500
1500
2500 o123 4 5 6
minutes
020.00
milli
Volts
22.50
500
1500
2500 191oC123 4 5 6
minutes
Figure 20.4. Part of the chromatogram showing the separation of six standard 18:1
isomers analysed at a column temperature of 191oC on a 120 m BPX-70 capillary
column
Peak identification: 1= t6, 2= t9, 3= t11, 4=c6, 5=c9, 6=c1
The chromatogram in Figure 20.1 shows that all six isomers could be identified with baseline
separation at a column temperature of 151oC. The difference in elution time (retention time)
between the first isomer and the last one was 6.21 minutes. This was enough time for all the
isomers to elute without overlapping, but the analyses took nearly 85 minutes. Using a
column temperature of 191oC, (Figure 20.4) the analysis time was shortened to just over 20
minutes and the time difference between the first and last peak decreased to 1 minute only.
Yet all the fatty acids could still be identified, though the three trans isomers and the cis-6 and
cis-9 isomers started overlapping. With a column temperature of 181oC, (Figure 20.3) all the
isomers were baseline separated with very little overlapping and the analysis time was
approximately 28 minutes.
68
Table 7. The effect of column temperature on the percentage composition of the different fatty acid isomers analysed with a 120 m BPX-70 capillary column
Isomers
Trans-6 Trans-9 Trans-11 Cis-6 Cis-9 Cis-11 Total
Temp.
151oC 17.5 14.4 22.9 11.4 18.7 15.1 100
171oC 17.6 14.3 23.0 11.3 18.7 15.1 100
181oC 17.6 14.3 23.0 11.3 18.7 15.1 100
191oC 17.9 14.0 23.1 10.8 19.1 15.1 100
Average 17.7 14.3 23.0 11.2 18.8 15.1
STD. 0.2 0.2 0.1 0.3 0.2 0.0
Table 7 shows that the percentage composition of the six isomers in the standard mixture
differs very little when using different column temperatures. It was only at a column
temperature of 191oC that the area percentages of the different isomers differed slightly from
the others. This could only be because of the overlapping of some of the peaks at the higher
column temperature. These results also show that the elution order of a standard mixture of
cis and trans FAME isomers, with different positional and geometrical structures do not
change with different column temperatures between 151o oC and 191 C and hydrogen gas flow
rates between 26 and 38 cm sec-1 using a 120 m BPX-70 capillary column. However, it shows
that higher column temperatures cause some peak overlapping. From Table 7 it is evident that
the concentration of the individual isomers stays constant at the three lower column
temperatures, as indicated by the relatively small standard deviation in their percentage
composition. It is concluded that a standard mixture of six different cis and trans FAME
isomers can be separated on a 120 m BPX-70 capillary column using any column temperature
between 151o oC and 181 C.
Previous literature (Kramer et al., 2004; Ratnayake et al., 2002; Aro et al., 1998; de Koning et
al., 2001) stated that column temperatures have a major effect on the separation of FAMEs
prepared from partially hydrogenated plant oils. Not only does the column temperature have
an effect on the retention times of the different fatty acid isomers, but also do some isomers
69
overlap when using a specific temperature. Changing the column temperature, however,
causes other isomers to overlap.
To optimise the column temperature for the analyses of margarine FAMEs, the prepared
pooled margarine sample was injected at different column temperatures.
4.5 Evaluation of different column temperatures
The nature and velocity of the carrier gas are primary considerations for the efficiency of a
given column. Hydrogen was used as the carrier gas, because of its high diffusivities and low
resistance to mass transfer (Christie, 1989). Different column gas flow rates between
26 cm sec-1 -1and 38 cm sec were also utilised. It was found that except for a change in the
retention times, there was very little effect on resolution, column efficiency and elution order
of the different isomers, when hydrogen was utilised. Thus, the precise flow rate was less
critical. Other authors also observed no loss of resolution, with different flow rates
(Ratnayake et al., 2006). This effect was also illustrated by a so-called Van Deemter plot of
the variation in the height of an effective theoretical plate with carrier gas velocities for
hydrogen, helium and nitrogen, where it could be seen that with hydrogen the height of the
effective theoretical plate varied little, with changes in the flow rate (Figure 12 on page 32).
The GLC results of the pooled margarine FAME sample, using column temperatures between
151o oC and 197 C, are demonstrated in the next twelve chromatograms (Figures 21.1-21.12).
With a column temperature of 151oC, (Figure 21.1) five peaks, 1-5 were separated before the
main cis-9, 18:1 fatty acid and six peaks, 6-11 after that. With column temperatures between
155o oC and 170 C, only four peaks could be separated before the cis-9, 18:1 fatty acid peak,
because peak 1 and 2 overlapped. At a column temperature of 170oC, a small peak (peak 12)
became separated from peak 10. With an increase in column temperature this peak slowly
moved towards peak 9 until it became part of peak 9 at a column temperature of 181oC. With
column temperatures between 175oC and 183oC, five peaks were again separated before the
main isomer as peak 13 became separated from the cis-9, 18:1 fatty acid peak. A new small
peak (peak 14) also became separated from peak 11, to give 8 identifiable peaks after cis-9,
18:1 peak. Raising the column temperature above 183oC, peaks 3 and 4 started to overlap and
peak 6 started to overlap with the main cis-9, 18:1 peak. At a column temperature of 197oC,
70
peaks 1 and 2 and peaks 3 and 4 overlapped and the main peak overlapped peak 6, leaving
only four identifiable peaks before the main isomer and six separated peaks thereafter.
0
100
300
500
700
00
oC
000
A B C
9 10 11000
oC
000
A C
12 34 5
6 7 8
80.000
95.00
oC
000
A C
9 10 11000
oC
00
75.0 85.0 90.00
900
A C
12 34 5
6 7 8
milli
Vol
ts
minutes
0
100
300
500
700
00
oC
000
A B C
9 10 11000
oC
000
A C
12 34 5
6 7 8
80.000
95.00
oC
000
A C
9 10 11000
151oC
00
75.0 85.0 90.00
900
A C
12 34 5
6 7 8
milli
Vol
ts
minutes
0
100
300
500
700
00
oC
000
A B C
9 10 11000
oC
000
A C
12 34 5
6 7 8
80.000
95.00
oC
000
A C
9 10 11000
oC
00
75.0 85.0 90.00
900
A C
12 34 5
6 7 8
milli
Vol
ts
minutes
0
100
300
500
700
00
oC
000
A B C
9 10 11000
oC
000
A C
12 34 5
6 7 8
80.000
95.00
oC
000
A C
9 10 11000
151oC
00
75.0 85.0 90.00
900
A C
12 34 5
6 7 8
milli
Vol
ts
minutes
0
100
300
500
700
00
oC
000
A B C
9 10 11000
oC
000
A C
12 34 5
6 7 8
80.000
95.00
oC
000
A C
9 10 11000
oC
00
75.0 85.0 90.00
900
A C
12 34 5
6 7 8
milli
Vol
ts
minutes
0
100
300
500
700
00
oC
000
A B C
9 10 11000
oC
000
A C
12 34 5
6 7 8
80.000
95.00
oC
000
A C
9 10 11000
151oC
00
75.0 85.0 90.00
900
A C
12 34 5
6 7 8
milli
Vol
ts
minutes
0
100
300
500
700
00
oC
000
A B C
9 10 11000
oC
000
A C
12 34 5
6 7 8
80.000
95.00
oC
000
A C
9 10 11000
oC
00
75.0 85.0 90.00
900
A C
12 34 5
6 7 8
milli
Vol
ts
minutes
0
100
300
500
700
00
oC
000
A B C
9 10 11000
oC
000
A C
12 34 5
6 7 8
80.000
95.00
oC
000
A C
9 10 11000
151oC
00
75.0 85.0 90.00
900
A C
12 34 5
6 7 8
milli
Vol
ts
minutes
Figure 21.1. Chromatogram of sample analysed at column temperature of 151o C
(11 peaks can be separated)
71
minutes
milli
Vol
ts
62.5 67.5 72.5 77.5 82.5
0
500
1000
1500
155oC
A B C
+
minutes
milli
Vol
ts
62.5 67.5 72.5 77.5 82.5
0
500
1000
1500
A B C
2 345
1
minutes
milli
Vol
ts
62.5 67.5 72.5 77.5 82.5
0
500
1000
1500
155oC
A B C
+
6 7 8
minutes
milli
Vol
ts
62.5 67.5 72.5 77.5 82.5
0
500
1000
1500
A B C
2 345
1
9 10 11
minutes
milli
Vol
ts
62.5 67.5 72.5 77.5 82.5
0
500
1000
1500
155oC
A B C
+
minutes
milli
Vol
ts
62.5 67.5 72.5 77.5 82.5
0
500
1000
1500
A B C
2 345
1
minutes
milli
Vol
ts
62.5 67.5 72.5 77.5 82.5
0
500
1000
1500
155oC
A B C
+
6 7 8
minutes62.5 67.5 72.5 77.5 82.5
0
500
1000
1500
A B C
2 3451
9 10 11
minutes
milli
Vol
ts
62.5 67.5 72.5 77.5 82.5
0
500
1000
1500
155oC
A B C
+
minutes
milli
Vol
ts
62.5 67.5 72.5 77.5 82.5
0
500
1000
1500
A B C
2 345
1
minutes
milli
Vol
ts
62.5 67.5 72.5 77.5 82.5
0
500
1000
1500
155oC
A B C
+
6 7 8
minutes
milli
Vol
ts
62.5 67.5 72.5 77.5 82.5
0
500
1000
1500
A B C
2 345
1
9 10 11
minutes
milli
Vol
ts
62.5 67.5 72.5 77.5 82.5
0
500
1000
1500
155oC
A B C
+
minutes
milli
Vol
ts
62.5 67.5 72.5 77.5 82.5
0
500
1000
1500
A B C
2 345
1
minutes
milli
Vol
ts
62.5 67.5 72.5 77.5 82.5
0
500
1000
1500
155oC
A B C
+
6 7 8
minutes62.5 67.5 72.5 77.5 82.5
0
500
1000
1500
A B C
2 3451
9 10 11
\Figure 21.2. Chromatogram of sample analysed at column temperature of 155oC
(10 peaks can be separated)
72
mill
iVol
tsm
illiV
olts
minutes47.5 50.0 52.5 55.0 57.5 60.0
0
500
1000
1500
A B C
160
+
minutes47.5 50.0 52.5 55.0 57.5 60.0
0
500
1000
1500
A B C
9 10 11
mill
iVol
tsm
illiV
olts
minutes47.5 50.0 52.5 55.0 57.5 60.0
0
500
1000
1500
A B C
o
2 3451
minutes47.5 50.0 52.5 55.0 57.5 60.0
0
500
1000
1500
A B C
C
6 7 8
mill
iVol
tsm
illiV
olts
minutes47.5 50.0 52.5 55.0 57.5 60.0
0
500
1000
1500
A B C
160
+
minutes47.5 50.0 52.5 55.0 57.5 60.0
0
500
1000
1500
A B C
9 10 11
mill
iVol
tsm
illiV
olts
minutes47.5 50.0 52.5 55.0 57.5 60.0
0
500
1000
1500
A B C
o
2 3451
minutes47.5 50.0 52.5 55.0 57.5 60.0
0
500
1000
1500
A B C
C
6 7 8
mill
iVol
tsm
illiV
olts
minutes47.5 50.0 52.5 55.0 57.5 60.0
0
500
1000
1500
A B C
160
+
minutes47.5 50.0 52.5 55.0 57.5 60.0
0
500
1000
1500
A B C
9 10 11
mill
iVol
tsm
illiV
olts
minutes47.5 50.0 52.5 55.0 57.5 60.0
0
500
1000
1500
A B C
o
2 3451
minutes47.5 50.0 52.5 55.0 57.5 60.0
0
500
1000
1500
A B C
C
6 7 8
mill
iVol
tsm
illiV
olts
minutes47.5 50.0 52.5 55.0 57.5 60.0
0
500
1000
1500
A B C
160
+
minutes47.5 50.0 52.5 55.0 57.5 60.0
0
500
1000
1500
A B C
9 10 11
mill
iVol
tsm
illiV
olts
minutes47.5 50.0 52.5 55.0 57.5 60.0
0
500
1000
1500
A B C
o
2 3451
minutes47.5 50.0 52.5 55.0 57.5 60.0
0
500
1000
1500
A B C
C
6 7 8
Figure 21.3. Chromatogram of sample analysed at column temperature of 160oC
(10 peaks can be separated)
73
74
Figure 21.4. Chromatogram of sample analysed at column temperature of 165oC
(10 peaks can be separated)
minutes
milli
Vol
ts
42.5 45.0 47.5 50.0 52.5 55.0
0
500
1000
1500
o
A B C
+
minutes
milli
Vol
ts
42.5 45.0 47.5 50.0 52.5 55.0
0
500
1000
1500
o
A B C
9 10 11
minutes
milli
Vol
ts
42.5 45.0 47.5 50.0 52.5 55.0
0
500
1000
1500
oC
A B C
1
minutes
milli
Vol
ts
42.5 45.0 47.5 50.0 52.5 55.0
0
500
1000
1500
165
A B C
2 34 5
6 7 8
minutes
milli
Vol
ts
42.5 45.0 47.5 50.0 52.5 55.0
0
500
1000
1500
o
A B C
+
minutes
milli
Vol
ts
42.5 45.0 47.5 50.0 52.5 55.0
0
500
1000
1500
o
A B C
9 10 11
minutes
milli
Vol
ts
42.5 45.0 47.5 50.0 52.5 55.0
0
500
1000
1500
oC
A B C
1
minutes
milli
Vol
ts
42.5 45.0 47.5 50.0 52.5 55.0
0
500
1000
1500
165
A B C
2 34 5
minutes
milli
Vol
ts
42.5 45.0 47.5 50.0 52.5 55.0
0
500
1000
1500
o
A B C
+
minutes
milli
Vol
ts
42.5 45.0 47.5 50.0 52.5 55.0
0
500
1000
1500
o
A B C
9 10 11
minutes
milli
Vol
ts
42.5 45.0 47.5 50.0 52.5 55.0
0
500
1000
1500
oC
A B C
1
minutes
milli
Vol
ts
42.5 45.0 47.5 50.0 52.5 55.0
0
500
1000
1500
165
A B C
2 34 5
6 7 8
minutes
milli
Vol
ts
42.5 45.0 47.5 50.0 52.5 55.0
0
500
1000
1500
o
A B C
+
minutes
milli
Vol
ts
42.5 45.0 47.5 50.0 52.5 55.0
0
500
1000
1500
o
A B C
9 10 11
minutes
milli
Vol
ts
42.5 45.0 47.5 50.0 52.5 55.0
0
500
1000
1500
oC
A B C
1
minutes
milli
Vol
ts
42.5 45.0 47.5 50.0 52.5 55.0
0
500
1000
1500
165
A B C
2 34 5
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
+
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
2345
1
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
o
6 7 8
9 10 11
12
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
170o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
+
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
2345
1
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
oC
6 7 8
9 10 11
12
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
+
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
2345
1
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
o
6 7 8
9 10 11
12
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
170o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
+
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
2345
1
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
oC
6 7 8
9 10 11
12
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
+
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
2345
1
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
o
6 7 8
9 10 11
12
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
170o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
+
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
2345
1
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
oC
6 7 8
9 10 11
12
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
+
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
2345
1
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
o
6 7 8
9 10 11
12
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
170o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
+
o
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
2345
1
minutes
milli
Vol
ts
37.5 40.0 42.5 45.0 47.5
0
500
1000
1500A B C
oC
6 7 8
9 10 11
12
Figure 21.5. Chromatogram of sample analysed at column temperature of 170oC
(11 peaks can be separated)
75
minutes
mill
iVol
ts
30.00 32.50 35.00 37.50
0
500
1000
1500
A B C
+1
minutes
mill
iVol
ts
30.00 32.50 35.00 37.50
0
500
1000
1500
o
minutes
mill
iVol
ts
30.00 32.50 35.00 37.50
0
500
1000
1500
35.00 37.50
0
500
1000
1500
oC
2 345
6 7 8
12
13
14
175
minutes
mill
iVol
ts
30.00 32.50 35.00 37.50
0
500
1000
1500
A B C
+1
minutes
mill
iVol
ts
30.00 32.50 35.00 37.50
0
500
1000
1500
o
minutes
mill
iVol
ts
30.00 32.50 35.00 37.50
0
500
1000
1500
35.00 37.50
0
500
1000
1500
oC
2 345
6 7 8
12
13
14
175
minutes
mill
iVol
ts
30.00 32.50 35.00 37.50
0
500
1000
1500
A B C
+1
minutes
mill
iVol
ts
30.00 32.50 35.00 37.50
0
500
1000
1500
o
minutes
mill
iVol
ts
30.00 32.50 35.00 37.50
0
500
1000
1500
35.00 37.50
0
500
1000
1500
oC
2 345
6 7 8
12
13
14
175
minutes
mill
iVol
ts
30.00 32.50 35.00 37.50
0
500
1000
1500
A B C
+1
minutes
mill
iVol
ts
30.00 32.50 35.00 37.50
0
500
1000
1500
o
minutes
mill
iVol
ts
30.00 32.50 35.00 37.50
0
500
1000
1500
35.00 37.50
0
500
1000
1500
oC
2 345
6 7 8
12
13
14
175
Figure 21.6. Chromatogram of sample analysed at column temperature of 175oC
(13 peaks can be separated)
76
minutes
30.00 32.50 35.00
0
500
1000
1500
177 CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
9 10 11
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
o
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
mill
iVol
ts
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
1
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
A B C
2345+
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
A B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
A B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
mill
iVol
ts
6 7 8
12
13
14
minutes30.00 32.50 35.00
0
500
1000
1500
177 CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
9 10 11
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
o
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
mill
iVol
ts
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
1
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
A B C
2345+
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
A B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
A B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
mill
iVol
ts
6 7 8
12
13
14
minutes30.00 32.50 35.00
0
500
1000
1500
177 CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
9 10 11
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
o
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
mill
iVol
ts
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
1
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
A B C
2345+
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
A B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
A B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
mill
iVol
ts
6 7 8
12
13
14
minutes30.00 32.50 35.00
0
500
1000
1500
177 CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
9 10 11
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
o
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
mill
iVol
ts
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
1
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
CA B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
A B C
2345+
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
A B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
minutes30.00 32.50 35.00
0
500
1000
1500
A B C
minutes30.00 32.50 35.00
0
500
1000
1500
C
mill
iVol
ts
6 7 8
12
13
14
Figure 21.7. Chromatogram of sample analysed at column temperature of 177oC
(13 peaks can be separated)
77
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
9 10 11
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C 1
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
2345
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
13
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
+
minutes27.50 30.00 32.50
0
500
1000
1500
oC
B C
6 7 8
12
14
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
9 10 11
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
179oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C 1
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
2345
minutes
mill
iVol
ts
27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
13
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
minutes27.50 30.00 32.50
0
500
1000
1500
oC
A B C
+
minutes27.50 30.00 32.50
0
500
1000
1500
oC
B C
6 7 8
12
14
Figure 21.8. Chromatogram of sample analysed at column temperature of 179oC
(13 peaks can be separated)
78
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
181oC
A B C
+
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
9 10 11
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
13
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
23451
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
6 7 8
14
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
181oC
A B C
+
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
9 10 11
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
13
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
23451
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
6 7 8
14
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
181oC
A B C
+
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
9 10 11
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
13
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
23451
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
6 7 8
14
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
181oC
A B C
+
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
9 10 11
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
13
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
23451
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
A B C
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
6 7 8
14
Figure 21.9. Chromatogram of sample analysed at column temperature of 181oC
(12 peaks can be separated)
79
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500A B C
2 345+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
om
illiV
olts
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
1
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
om
illiV
olts
6 7 8
14
13
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
9 10 11
183o
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
om
illiV
olts
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
Cm
illiV
olts
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
2 345+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
om
illiV
olts
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
1
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
om
illiV
olts
14
13
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
9 10 11
183o
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
om
illiV
olts
minutes25.00 27.50 30.00
0
500
1000
1500
2000
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
o
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
minutes25.00 27.50 30.00
0
500
1000
1500
2000
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
minutes25.00 27.50 30.00
0
500
1000
1500
2000
minutes25.00 27.50 30.00
0
500
1000
1500
2000
minutes25.00 27.50 30.00
0
500
1000
1500
2000
Cm
illiV
olts
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500A B C
2 345+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
om
illiV
olts
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
1
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
om
illiV
olts
6 7 8
14
13
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
9 10 11
183o
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
om
illiV
olts
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
Cm
illiV
olts
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
2 345+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
om
illiV
olts
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
1
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
om
illiV
olts
14
13
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
2500
9 10 11
183o
2500
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
om
illiV
olts
minutes25.00 27.50 30.00
0
500
1000
1500
2000
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
o
+
minutes25.00 27.50 30.00
0
500
1000
1500
2000
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
minutes25.00 27.50 30.00
0
500
1000
1500
2000
o
minutes25.00 27.50 30.00
0
500
1000
1500
2000
minutes25.00 27.50 30.00
0
500
1000
1500
2000
minutes25.00 27.50 30.00
0
500
1000
1500
2000
minutes25.00 27.50 30.00
0
500
1000
1500
2000
Cm
illiV
olts
Figure 21.10. Chromatogram of sample analysed at column temperature of 183oC
(12 peaks can be separated)
80
minutes
22.50 25.000
500
1000
1500
oC
A B c
24 5
minutes22.50 25.00
0
500
1000
1500
o190m
illiV
olts
mill
iVol
ts
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
o190
9 10 11
mill
iVol
tsm
illiV
olts
minutes22.50 25.00
0
500
1000
1500
oC
+
minutes22.50 25.00
0
500
1000
1500
minutes22.50 25.00
0
500
1000
1500
oC
+
minutes22.50 25.00
0
500
1000
1500
o190
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
o190m
illiV
olts
mill
iVol
ts
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
o190m
illiV
olts
mill
iVol
ts
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
o190
1 3
6 7 8
14
13
minutes22.50 25.00
0
500
1000
1500
oC
24 5
minutes22.50 25.00
0
500
1000
1500
o190m
illiV
olts
mill
iVol
ts
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
o190
9 10 11
mill
iVol
tsm
illiV
olts
minutes22.50 25.00
0
500
1000
1500
oC
+
minutes22.50 25.00
0
500
1000
1500
minutes22.50 25.00
0
500
1000
1500
oC
+
minutes22.50 25.00
0
500
1000
1500
o190
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
o190m
illiV
olts
mill
iVol
ts
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
o190m
illiV
olts
mill
iVol
ts
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
o190
1 3
6 7 8
14
13
minutes22.50 25.00
0
500
1000
1500
oC
A B c
24 5
minutes22.50 25.00
0
500
1000
1500
o190m
illiV
olts
mill
iVol
ts
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
o190
9 10 11
mill
iVol
tsm
illiV
olts
minutes22.50 25.00
0
500
1000
1500
oC
+
minutes22.50 25.00
0
500
1000
1500
minutes22.50 25.00
0
500
1000
1500
oC
+
minutes22.50 25.00
0
500
1000
1500
o190
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
o190m
illiV
olts
mill
iVol
ts
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
o190m
illiV
olts
mill
iVol
ts
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
o190
1 3
6 7 8
14
13
minutes22.50 25.00
0
500
1000
1500
oC
24 5
minutes22.50 25.00
0
500
1000
1500
o190m
illiV
olts
mill
iVol
ts
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
o190
9 10 11
mill
iVol
tsm
illiV
olts
minutes22.50 25.00
0
500
1000
1500
oC
+
minutes22.50 25.00
0
500
1000
1500
minutes22.50 25.00
0
500
1000
1500
oC
+
minutes22.50 25.00
0
500
1000
1500
o190
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
o190m
illiV
olts
mill
iVol
ts
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
o190m
illiV
olts
mill
iVol
ts
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
minutes22.50 25.00
0
500
1000
1500
oC
minutes22.50 25.00
0
500
1000
1500
o190
1 3
6 7 8
14
13
Figure 21.11. Chromatogram of sample analysed at column temperature of 190oC
(11 peaks can be separated)
81
minutes
milli
Vol
ts
0
500
1000
1500
197oC
A B C
+
minutes
milli
Vol
ts
17.50 18.50 19.50 20.50 21.50
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
+
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
17.50 18.50 19.50 20.50 21.50
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
2413
7 8
910 11
5
13
14
minutes
milli
Vol
ts
0
500
1000
1500
197oC
+
minutes
milli
Vol
ts
17.50 18.50 19.50 20.50 21.50
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
+
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
17.50 18.50 19.50 20.50 21.50
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
2413
7 8
5
14
minutes
milli
Vol
ts
0
500
1000
1500
197oC
A B C
+
minutes
milli
Vol
ts
17.50 18.50 19.50 20.50 21.50
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
+
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
17.50 18.50 19.50 20.50 21.50
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
2413
7 8
910 11
5
13
14
minutes
milli
Vol
ts
0
500
1000
1500
197oC
+
minutes
milli
Vol
ts
17.50 18.50 19.50 20.50 21.50
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
+
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
17.50 18.50 19.50 20.50 21.50
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
minutes
milli
Vol
ts
0
500
1000
1500
197oC
2413
7 8
5
14
Figure 21.12. Chromatogram of sample analysed at column temperature of 197oC
(10 peaks can be separated)
82
A summary of the peaks separated at the different column temperatures, as well as the
temperature effect on peak areas, because of overlapping, are demonstrated in Table 5.
Table 8. The percentage composition of the different peaks and of the main 18:1 fatty acid isomer in the pooled margarine sample, using different column temperature between 151o oC and 197 C
Peak 1 2 1+2 3 4 3+4 5 13 18:1 6 7 8 9 10 11 12 14151oC 2.9 4.1 - 4.6 4.0 - 2.6 - 25.4 1.1 1.7 2.4 0.3 0.3 0.3 - -155oC - - 7.1 4.6 4.0 - 2.8 - 25.1 0.9 1.7 2.4 0.4 0.3 0.3 - -160oC - - 7.0 4.6 3.7 - 3.2 - 25.2 0.8 1.4 2.5 0.3 0.4 0.4 - -165oC - - 7.1 4.6 3.7 - 3.1 - 25.1 1.0 1.4 2.5 0.4 0.3 0.3 - -170oC - - 7.2 4.7 3.8 - 2.0 - 25.3 1.1 1.5 2.5 0.4 0.2 0.2 0.1 -175oC - - 7.1 4.7 3.7 - 2.6 1.7 24.4 0.6 1.4 2.5 0.3 0.1 0.2 0.1 0.1177oC - - 7.1 4.7 3.6 - 2.6 1.9 24.2 0.5 1.4 2.5 0.4 0.1 0.3 0.1 0.1179oC - - 7.1 4.6 3.6 - 2.5 2.1 24.1 0.6 1.4 2.6 0.4 0.1 0.3 0.1 0.2181oC - - 7.0 4.7 3.6 - 2.5 2.2 24.0 0.5 1.4 2.6 0.5 0.3 0.1 - 0.2183oC - - 7.1 4.8 3.5 - 2.5 2.3 23.9 0.5 1.4 2.6 0.5 0.3 0.1 - 0.3190oC - - 7.0 - - 8.5 2.3 2.3 24.6 0.2 1.2 2.6 0.5 0.3 0.1 - 0.3197oC - - 7.1 - - 8.5 2.3 2.4 25.1 - 0.9 2.6 0.6 0.3 0.1 - 0.3
The differences in the percentage areas of some of the peaks, at different temperatures, could
only be because of overlapping, since there were no significant differences in the percentage
areas of the principal 16:0, 18:0 and 18:2 fatty acids peaks. Although, depending on the
column temperature, between 10 and 13 different peaks could be separated. This is still less
than the reported number of isomers that could be found in partially hydrogenated plant oils.
The literature states that the octadecenoic acid (18:1) isomer group, which is the primary
isomeric fatty acid group in partially hydrogenated plant oils, can constitute up to 26 different
cis and trans isomers (Ratnayake et al., 2002). Kramer et al. identified 12 different trans fatty
acid isomers in a partially hydrogenated oil sample, although the peak areas of some of these
peaks were very small (Kramer et al., 2004). Aro et al. identified 13 isomers using a 100-m
CP-Sil capillary column with a column temperature of 170oC. Preceding Ag-TLC, improved
their separation to 16 isomers (Aro et al., 1998).
83
It was concluded that even with a highly-polar 120 m BPX-70 column, it was not possible to
separate all the different isomers in the margarine samples using only one isothermal column
temperature. Using temperature programming will not solve this problem, because the
retention times of the different cis and trans isomers are very close to one another, leaving
very little time for temperature programming. It has been reported that temperature
programming gives less satisfactory separation of cis- and trans- 18:1 isomers, whereas
optimal isothermal column conditions provide far more improved separation of the 18:1
isomers, although some isomers will always overlap (Ratnayake et al., 2002). Ideally, each
sample will have to be injected at different column temperatures a number of times. This will
be very time consuming and still not guarantee the separation and identification of all the cis
and trans isomers in the samples.
The chromatograms (Figures 21.1-21.12) and the evaluation of the percentage area counts of
the different peaks (Table 8), demonstrated that between 175o oC and 183 C the most individual
peaks could be identified, although not all the peaks were baseline separated. At a column
temperature of 175oC, peak 13 became separated from the main isomer and the separation
improved with a raise in column temperature. On the other hand, at 175oC, peak 6 was still
separated, though with a rise in temperature it started to overlap the main cis-9, 18:1 isomer.
Decreasing the column temperature to 151oC, resulted in a better separation between the early
eluting trans isomers, but the later eluting isomers were not well resolved and the analytical
time was very long. The average sum of all the isomers, between the different column
temperatures, was 49.6 ± 0.3%. Therefore, choosing a column temperature, which gives the
highest total percentage of all the different isomers in a sample, will be of little use, because
the average total concentration of the isomers between the different column temperatures was
nearly the same. Therefore, it was decided to use a column temperature of 181oC for the
identification and quantification of the different isomers in the margarine samples, since the
main cis-9, 18:1 fatty acid overlapped the fewest isomers at this temperature, and the
analytical time was reasonable.
The identification of the individual fatty acid methyl ester peaks was done by comparing the
retention times of the sample peaks with those obtained from cis and trans FAME standards
(Nu- Chek- Prep, INC. Elysian, Minnesota, USA). These were analysed under exactly the
same analytical conditions as used for the sample. To help with the identification of
84
overlapping peaks, the sample was spiked with known FAME standard isomers. To further
assist with the identification, equivalent chain-length (ECL) values for cis and trans FAMEs
from SGE were used (www.sge.com). The quantification of the different fatty acids was done
with the 17:0 internal standard added to the sample.
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
181oC
A B C
+
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
9 10 11
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
13
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
23451
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
6 7 8
14
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
181oC
A B C
+
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
9 10 11
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
13
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
23451
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
6 7 8
14
Figure 21.13. Chromatogram of the pooled sample analysed at column temperature oC of 181
In the 181oC chromatogram (Figure 21.13) the first peak was identified by spiking to be at
least the unresolved trans-6 and trans-9 isomers. This overlapping was confirmed in the
chromatogram of the analyses at a column temperature of 151oC, (Figure 21.1) where it was
confirmed to be two peaks representing at least two isomers. Trans-7 and -8 could not be
identified, but the literature states that the trans-6, -7 and -8 isomers are always eluting as one
peak (Precht et al., 1996; Ratnayake et al., 2002), thus peak 1 was identified as the unresolved
trans-6, -7, -8 and -9 isomers. Peaks 3, 4 and 5 were identified as trans-10, -11 and -12. Peak
13 was identified as the unresolved trans-13 and -14 isomers, because using a highly-polar,
85
very long capillary column, the trans-13 and -14 isomers always elute as one peak (Precht et
al., 1996; Ratnayake et al., 2002). The peak marked B is the main cis-9, 18:1 isomer. The cis-
6, -7 and -8 isomers that normally elute before the cis-9, could not be separated. After spiking
the sample with a cis-6 FAME standard, it was found that the cis-6 isomer forms part of the
unresolved trans-13 and -14 isomers’ peak. This observation was confirmed by Ratnayake et
al., who found that the unresolved trans-13 and -14 isomers’ peak always overlaps the cis-6, -
7 and -8 isomers when using a long highly-polar column. These 3 cis isomers are of minor
importance in the analyses of partially hydrogenated plant oil samples, because they contained
less than 0.2% of these three cis isomers (Ratnayake et al., 2002). The trans-15 isomer could
not be identified, and it was calculated that its retention time was very close to the retention
time of the cis-9, 18:1 isomer, and that this large peak is probably overlapping it.
Quantitatively, this is not a major drawback, because the trans-15 isomer is a minor
component in partially hydrogenated plant oils (Glew et al., 2006; Ratnayake et al., 2002).
Peaks 6-11 were identified as cis-10, -11, -12, -13, -14 and -15, respectively. Another
noteworthy improvement with a 120 m BPX-70 column was the partial separation of trans-16
(peak 14) and cis-15 (peak 11). The trans-16 isomer could only be resolved at an isothermal
column temperature above 175oC. Normally this isomer appeared as a shoulder on the edge of
the cis-15 isomer peak (Ratnayake et al., 2002). Except for the cis-16 isomer, all the other
peaks were identified.
Table 9 gives the results of the identification and quantification of the different peaks
analysed at 151oC, 170oC, 181o oC and 197 C. Although, some of the trans isomers eluted as a
group as well as some overlapping between cis and trans isomers, elaidic acid (trans-9, 18:1)
which is the major man-made trans fatty acid found in partially hydrogenated plant oils and
processed foods (Belury, 2002) and vaccenic acid (trans-11, 18:1) that occurs naturally in
foods from animal sources (Wolff, 1995), could be identified. The two saturated fatty acids,
16:0 and 18:0, and the one polyunsaturated fatty acid, 18:2, in the sample were well resolved
with column temperatures between 151o oC and 197 C. The average concentrations of these
fatty acids between the different column temperatures were 16:0; 14.7 ± 0.2 mg/100 mg, 18:0;
8.3 ± 0.1 mg/100 mg and the 18:2; 19.6 ± 0.2 mg/100 mg. The concentration range of the
main cis-9, 18:1 fatty acid isomer, between the different column temperatures was between
20.4 and 21.5 mg/100 mg. This large difference was because of the overlapping with peak 13
at a lower temperature, and peak 6 at a temperature of 197oC. The total fatty acid
86
87
concentration of the margarine, as determined at a column temperature of 181oC, was 85.2
mg/100 mg. This fatty acid concentration is the highest of the four column temperatures that
were evaluated, indicating that selecting a column temperature of 181oC for the analyses of
the samples is correct. The total trans fatty acid concentration, at a column temperature of
181oC, was 17.3%. The cis-6, -7, and -8 isomers that possibly overlapped with the trans-13
and -14 were calculated as trans isomers. The trans-15 isomer, which the cis-9 isomer
overlapped, was calculated as cis-9, 18:1.
Table10 gives the results of the pooled sample that was injected five times into a 120 m BPX-
70 capillary column with a column temperature of 181oC. The total fatty acid concentration of
the sample was 84.8 ± 0.2 mg/100 mg and the average concentration of the major trans
isomers (trans-6 to trans-14) was 17.2 ± 0.2 mg/100 mg. The low standard deviation of the
individual saturated and polyunsaturated fatty acids, as well as the different cis and trans 18:1
isomers indicate good reproducibility. The percentage coefficient of variance was also less
than 2% for the major cis and trans isomers, except for those with low concentrations where
this statistical measurement is not a good tool.
Table 9. The total fatty acid concentration in mg/100 g of the pooled margarine sample injected into a 120 m BPX-70
capillary column at different column temperatures between 151o oC and 197 C
16:0 18:0 18:1 18:2 Totalt6,t7 t9 t6,t7 t10 t11 t10,t11 t12 t13,t14 c9 c9,c10 c10 c11 c12 c13 c14 c15 t16
t8 t8,t9 c6,c7,c8 t15 t15151oC 15.0 8.4 2.4 3.5 - 3.9 3.4 - 2.2 - 21.5 - 0.9 1.4 2.0 0.3 0.3 0.3 - 19.6 85.1170oC 14.7 8.3 - - 6.0 3.9 3.2 - 1.6 - 21.1 - 0.9 1.2 2.1 0.3 0.2 0.3 - 19.6 83.5181oC 14.8 8.3 - - 6.0 4.0 3.1 - 2.1 1.9 20.4 - 0.4 1.2 2.2 0.5 0.2 0.2 0.2 19.8 85.2197oC 14.5 8.3 - - 6.0 - - 7.1 1.9 1.9 - 21.1 - 0.8 2.2 0.5 0.3 0.2 0.3 19.7 84.9
AVG 14.8 8.3 - - 6.0 3.9 3.2 - 2.0 1.9 21.0 - 0.7 1.1 2.1 0.4 0.3 0.3 - 19.7 84.7STD 0.2 0.0 - - 0.0 0.1 0.2 - 0.2 0.0 0.6 - 0.3 0.3 0.1 0.1 0.1 0.1 - 0.1 0.8
88
Table 10. The total fatty acid concentration in mg/100 mg of the pooled margarine sample injected five times into a 120 m
16:0 18:0 18:1 18:2 Totalt6,t7 t10 t11 t12 t13,t14 c9 c10 c11 c12 c13 c14 c15 t16t8,t9 c6,c7,c8 t15
A 14.7 8.3 6.0 3.9 3.1 2.2 1.7 20.4 0.5 1.2 2.1 0.3 0.2 0.2 0.2 19.8 84.9B 14.8 8.3 6.0 4.0 3.1 2.2 1.6 20.8 0.5 1.2 2.2 0.3 0.2 0.2 0.3 19.8 84.9C 14.7 8.3 6.1 3.9 3.1 2.2 1.8 20.7 0.5 1.2 2.2 0.3 0.2 0.3 0.2 19.8 84.9D 14.8 8.3 6.0 4.0 3.1 2.1 1.9 20.6 0.4 1.2 2.2 0.5 0.2 0.3 0.2 19.8 85.1E 14.5 8.2 6.1 4.0 3.0 2.2 2.0 20.6 0.4 1.1 2.2 0.5 0.2 0.3 0.2 19.5 84.4
AVG 14.7 8.3 6.0 4.0 3.1 2.2 1.8 20.6 0.4 1.2 2.2 0.4 0.2 0.3 0.2 19.7 84.8STD 0.1 0.1 0.0 0.0 0.1 0.0 0.2 0.1 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.2 0.2%CV 0.8 0.7 0.8 1.0 1.9 1.8 8.4 0.7 7.8 2.7 1.1 21.1 6.1 5.0 20.3 0.8 0.3
oC BPX-70 capillary column at a column temperature of 181
89
4.6 Results of silver ion thin layer chromatography
Figure 22.1 shows a photograph of the developed Ag-TLC plate. The trans (1) and cis (2)
mono-unsaturated fatty acid fractions were well separated with no overlapping between the
two fractions. The fractions were analysed using the same GLC conditions that was used to
identify and quantify the unfractioned pooled sample.
1
23
4
1
23
4
Figure 22.1. Photograph of a TLC plate impregnated with 10% (w/v) silver nitrate
showing the separation of the cis and trans monounsaturated FAME isomer fractions in
the pooled margarine sample 1= Trans isomers. 2= Cis isomers. 3= Trans-9, 18:1 standard. 4= Cis-9, 18:1 standard
90
Part of the GLC chromatogram of the trans mono- unsaturated fatty acid fraction is given in Figure
22.2.
minutes
milli
Vol
ts
25.00 27.50 30.00-100
0
100
200
300
400
500
600
700
800
minutes
milli
Vol
ts
25.00 27.50 30.00-100
0
100
200
300
400
500
600
700
800
1
4
76
181oC
minutes
milli
Vol
ts
25.00 27.50 30.00-100
0
100
200
300
400
500
600
700
800
minutes
milli
Vol
ts
25.00 27.50 30.00-100
0
100
200
300
400
500
600
700
800
1
2
76
181 C
5
3
minutes
milli
Vol
ts
25.00 27.50 30.00-100
0
100
200
300
400
500
600
700
800
minutes
milli
Vol
ts
25.00 27.50 30.00-100
0
100
200
300
400
500
600
700
800
1
4
76
181oC
minutes
milli
Vol
ts
25.00 27.50 30.00-100
0
100
200
300
400
500
600
700
800
minutes
milli
Vol
ts
25.00 27.50 30.00-100
0
100
200
300
400
500
600
700
800
1
2
76
181 C
5
3
Figure 22.2. Part of the chromatogram of the trans mono-unsaturated FA fraction of the
pooled sample, after Ag-TLC separation. The fraction was analysed with a 120 m BPX-
70 capillary column at a column temperature of 181oC Peak identification: 1= trans- 6, 7, 8 and 9, 2=trans -10, 3= trans- 11, 4= trans- 12,
5= trans-13 and 14, 6= trans-15 and 7=trans-16
Seven different peaks could be identified from the chromatogram of the trans fraction (Figure
22.2). The trans-6, -7, -8 and -9 isomers still eluted as one peak (peak 1). Peaks 2-4 were
identified as trans-10, -11 and -12. Peak 5 is the unresolved trans-13 and -14 isomers. Peak 6
is the trans-15 isomer that is normally overlapped in the unfractioned samples by the main
cis-9 isomer, and peak 7 is the trans-16 isomer. None of the known resolved cis isomers were
present in this fraction, proving that Ag-TLC separated the cis and trans mono-unsaturated
fatty acid isomers completely, but the different overlapping trans isomers were still not
separated. This overlapping effect of the trans isomers was also reported by other authors
(Precht et al., 1996; Aro et al., 1998).
91
The GLC results (in percentage compositions) of the different trans isomers, with and without
fractionation with Ag-TLC, are summarised in Table. 11.
Table 11. The GLC results (in percentage composition) of the different trans 18:1 isomers, with and without preceding Ag-TLC separation.
Isomers Trans Trans
10 Trans
11 Trans
12 Trans 13+14
Trans 15
Trans 16
Total (%) 6 - 9
Method
Without Ag-TLC 33.7 23.9 19.4 13.2 10.4 -- 0.4 100
With 32.0 24.0 19.2 13.3 8.8 2.2 0.5 100 Ag-TLC
The same trans overlapping groups and individual isomers, with the exception of trans-15,
were identified. The difference of 1.7% between the “without Ag-TLC” and the “with Ag-
TLC” groups of the trans-6 to trans-9 isomeric group is likely because of the overlapping
effect of the cis-5 isomer that was identified in the cis mono-unsaturated fatty acid fraction
after Ag-TLC. The other noteworthy difference between the two methods was the well-
resolved trans-15 isomer, after Ag-TLC. With its 2.2 % composition of the total trans fatty
acids it could be seen that this trans isomer is not as negligible as some authors have reported
(Ratnayake et al., 2002). There is very little difference in the percentage compositions of the
trans-10, -11 and -12 isomers between the two methods. This is an indication that there are no
cis isomers eluting with the trans-10, -11 and -12 isomers in the unfractioned sample.
The chromatogram (Figure 22.3) of the cis mono-unsaturated fatty acid fraction showed 11
different peaks.
92
minutes
milli
Vol
ts
25.00 27.50 30.000
100
200
300
400
500
600
700
800
900
1 2 3 4
8
minutes25.00 27.50 30.000
100
200
300
400
500
600
700
800
900
1 2 3 4
181oC
minutes25.00 27.50 30.000
100
200
300
400
500
600
700
800
900
1 2 3 4
minutes25.00 27.50 30.000
100
200
300
400
500
600
700
800
900
1 2 3 49 10 11
oC
5
76
minutes
milli
Vol
ts
25.00 27.50 30.000
100
200
300
400
500
600
700
800
900
1 2 3 4
8
minutes25.00 27.50 30.000
100
200
300
400
500
600
700
800
900
1 2 3 4
181oC
minutes25.00 27.50 30.000
100
200
300
400
500
600
700
800
900
1 2 3 4
minutes25.00 27.50 30.000
100
200
300
400
500
600
700
800
900
1 2 3 49 10 11
oC
5
76
Figure 22.3. Part of the chromatogram of the cis mono-unsaturated fraction of the
pooled sample, after Ag-TLC separation. The fraction was analysed with a 120 m BPX-
70 capillary column at a column temperature of 181oC Peak identification: 1= cis-5, 2=cis-6, 3=cis-7, 4=cis-8, 5= cis-9, 6= cis-10, 7=cis-11, 8=cis-12, 9=cis-13,
10=cis-14 and 11=cis-15
The well-resolved peak 1 has not been mentioned in the most recent and prominent
publications on trans fatty acids in partially hydrogenated vegetable oils (Precht et al., 1996;
Ratnayake et al., 2002; Glew et al., 2006; Kramer et al., 2004; de Koning et al., 2001;
Ratnayake et al., 2006; Ledoux et al., 2000). From this peak’s retention time, it must elute
together with the overlapping group of trans-6 to trans-9 isomers in the unfractioned samples.
In the fractioned sample, this isomer elutes just before the known group of cis-6 to cis-8
isomers and, therefore, the assumption can be made that it is probably a cis-5 isomer.
However, it could not be proven, because a cis-5 standard was not available to confirm this.
To exclude the possibility of a contaminant, the extraction and separation was repeated with
similar results. Peaks 2, 3 and 4 were identified as the cis-6, -7 and -8 isomers that are
normally overlapped by the trans-13 and -14 isomers. The separation of these three cis
isomers, although not very good, was noted. In the trans mono-unsaturated fatty acid fraction,
it was not possible to separate any of the known overlapping trans groups. Peaks 5, 6, 7, 8, 9,
10 and 11 were identified as the cis-9 to cis-15 isomers.
93
Table 12. The GLC results (in percentage composition) of the different cis 18:1 isomers, with and without preceding Ag-TLC separation
Isomers Cis5
Cis6
Cis 7
Cis 8
Cis 9
Cis 10
Cis 11
Cis 12
Cis 13
Cis 14
Cis 15
Total (%) Method
Without -- -- -- -- 81.4 1.6 4.8 8.9 1.6 0.7 1.0 100 Ag-TLC
With 1.6 0.3 0.2 0.4 78.6 2.2 5.2 9.1 1.4 0.5 0.5 100 Ag-TLC
Table 12 gives the percentage composition of the cis isomers determined by the two methods.
With Ag-TLC separation, all the cis isomers that eluted before the main cis-9 isomer could be
identified, but were overlapped by the trans isomers in the unfractioned samples. The
difference in the percentage composition between the cis-9 isomer in the two methods was
caused by the overlapping effect of the trans-15 isomer in the unfractioned sample. With Ag-
TLC separation, the trans-15 did not form part of the cis-9 isomer and, therefore, its
percentage composition was 2.8% lower. With the exception of the cis-5 to cis-8 isomers, no
other new isomers were separated. The 0.5% difference in the composition of the cis-15
between the two methods is an indication that there was some overlapping with the trans-16
isomer in the unfractioned sample. Except for these few discrepancies, the percentage
composition of the other isomers between the two methods was very comparable.
The different isomers in the two fractions were not quantified, because Ag-TLC was only
used to separate the cis and trans isomeric groups and to see to what extent overlapping of
these isomers can influence the concentration of the different isomers in the unfractioned
samples. To quantify the different isomers, two internal standards with known concentrations,
a cis mono-unsaturated and a trans mono-unsaturated fatty acids, which are not occurring in
the samples, must be used. Unfortunately, with partial hydrogenation any number of different
isomers can be formed, making it impossible to predict which isomers do not occur in the
sample. Another method of quantifying the different isomers is to spike the sample by
pipetting precisely two aliquots of the sample extract into two transmethylation tubes. Known
concentrations of a cis-9, 18:1 standard, and a trans-9, 18:1 standard, should be added to one
94
of the tubes. These two standard isomers are used because they are normally the main cis and
trans isomers in partially hydrogenated oil samples. These duplicate samples are then
analysed and from the differences in area counts of the cis-9, 18:1 and the trans-9, 18:1 peaks
in the two aliquots, the concentration of the unspiked peak can be calculated. If the
concentration of one peak is known, the concentration of all the other peaks can be calculated.
However, Ag-TLC separations are laborious and to duplicate each sample is impractical for
routine analyses.
4.7 Results of the margarine samples
After evaluation and standardisation of the method, the margarine samples were analysed
using a 120 m BPX-70 capillary column at a column temperature of 181oC and a hydrogen
gas flow of 30 cm sec-1.
The results (Table 13) show that only two of the samples (Sample D (18.4 mg100mg) and
Sample J (2.3 mg/100 mg)) have a trans fatty acid concentration higher than 2% of the total
fatty acids. Sample D was exceptionally high with a total trans content of 18.4 %. Samples A,
B, C, E, G, H, I and K have also traces of trans fatty acids. Although the trans fatty acid
concentration is very low it still indicates the presence of partially hydrogenated vegetable oil,
because unhydrogenated vegetable oils have no trans fatty acids. The mean trans fatty acid
content of the 18 margarines was 1.3 mg/100 mg fatty acids. This is much lower than the
mean levels of 16.4 ± 2.6 mg/100mg reported for New Zealand (Lake et al, 1996) and 9.7
mg/100mg (standard deviation was not given) for the Czech Republic (Brat et al, 2000).
Except for samples D and J that obviously contain partially hydrogenated vegetable oil, none
of the other samples contained trans-11 isomers. This indicates that the margarines contain no
animal fats, because trans-11, 18:1 is the main naturally occurring trans isomer in animal fat
and milk (Sommerfeld, 1983). Samples B, C and N have high concentrations of palmitic acid
(16:0) and oleic acid (18:1) and low linoleic acid (18:2) indicating that they are most probably
manufactured from palm oil (Oils and Fats, 2005). The total fatty acid concentration in the 18
margarine samples varied between 44.1 and 98.3 mg/100 mg. This is directly related to the
total fat content of the samples. From the fat content of the samples only three of these can be
classified as margarines, because South African regulations require that margarine must at
least contain 80% fat (Draft Regulations Relating to Labelling and Advertising of Foodstuffs
95
96
2002, no. R 1055). The rest of the samples, with less than 80% fat, are classified as “Table
spreads”. To manufacture a “Lite” margarine (Sample M) that normally has a lower fat
content, the manufacturers substitute some plant oils and fats with water. These “Lite”
margarines are normally softer, indicating higher polyunsaturated fatty acid content.
12:0 14:0 16:0 16:1 18:0 18:1 18:2 Total t6,t9 t10 t11 t12 t13,t14 c9 c10 c11 c12 c13 c14 c15 t16
(t7,t8)* (c6,c7,c8)* (t15)* A 2.9 1.4 15.2 0.1 2.7 0.1 0.0 0.0 0.0 0.0 20.7 0.0 0.6 0.0 0.0 0.0 0.0 0.0 18.9 62.6 B 0.2 0.9 40.2 0.1 4.0 0.1 0.0 0.0 0.0 0.0 32.6 0.0 0.5 0.0 0.0 0.0 0.0 0.0 8.1 86.8 C 0.3 1.2 49.9 0.1 5.1 0.1 0.0 0.0 0.0 0.0 33.1 0.0 0.6 0.0 0.0 0.0 0.0 0.0 8.0 98.3 D 0.0 0.2 16.1 0.0 8.7 6.3 4.2 3.2 2.3 2.1 21.5 0.5 1.2 2.3 0.5 0.3 0.3 0.3 20.4 90.4 E 2.2 1.2 18.9 0.1 2.6 0.1 0.0 0.0 0.0 0.0 18.5 0.0 0.4 0.0 0.0 0.0 0.0 0.0 11.1 55.0 F 2.4 1.4 24.1 0.1 2.5 0.0 0.0 0.0 0.0 0.0 21.0 0.0 0.4 0.0 0.0 0.0 0.0 0.0 5.3 57.2 G 4.9 2.3 16.9 0.1 3.8 0.1 0.0 0.0 0.0 0.0 16.5 0.0 0.3 0.0 0.0 0.0 0.0 0.0 9.4 54.3 H 5.1 2.6 23.1 0.1 5.4 0.1 0.0 0.0 0.0 0.0 23.6 0.0 0.4 0.0 0.0 0.0 0.0 0.0 13.7 74.0 I 1.0 1.1 16.7 0.0 2.4 0.6 0.4 0.0 0.1 0.0 16.5 0.0 0.3 0.1 0.0 0.0 0.0 0.0 5.0 44.1 J 2.6 1.8 25.9 0.1 3.7 1.3 0.5 0.3 0.2 0.0 26.1 0.0 0.5 0.1 0.0 0.0 0.0 0.0 8.4 71.6 K 4.7 3.1 25.2 0.1 3.8 0.1 0.0 0.0 0.0 0.0 23.6 0.0 0.4 0.0 0.0 0.0 0.0 0.0 8.5 69.5 L 1.7 1.0 16.9 0.1 2.3 0.0 0.0 0.0 0.0 0.0 15.3 0.0 0.3 0.0 0.0 0.0 0.0 0.0 7.9 45.3 M 1.3 0.9 7.1 0.0 6.2 0.0 0.0 0.0 0.0 0.0 14.9 0.0 0.3 0.0 0.0 0.0 0.0 0.0 36.0 66.8 N 0.1 0.8 30.8 0.1 3.4 0.0 0.0 0.0 0.0 0.0 30.6 0.0 0.5 0.0 0.0 0.0 0.0 0.0 9.9 76.3 O 2.2 1.7 21.0 0.1 4.0 0.0 0.0 0.0 0.0 0.0 21.2 0.0 0.4 0.0 0.0 0.0 0.0 0.0 14.8 65.3 P 2.1 1.4 15.0 0.1 2.7 0.0 0.0 0.0 0.0 0.0 20.5 0.0 0.6 0.0 0.0 0.0 0.0 0.0 18.7 61.2 Q 4.1 2.2 22.2 0.1 3.2 0.0 0.0 0.0 0.0 0.0 21.4 0.0 0.4 0.0 0.0 0.0 0.0 0.0 9.9 63.4 R 0.9 1.0 24.7 0.1 3.1 0.0 0.0 0.0 0.0 0.0 25.8 0.0 0.5 0.0 0.0 0.0 0.0 0.0 10.3 66.3
Table 13. The total fatty acid composition (mg/ 100 mg) of the margarine samples analysed with a 120 m BPX-70 capillary -1
column at a column temperature of 181oC and a hydrogen gas flow rate of 30 cm sec
* The isomers in the brackets cannot be identified because of overlapping
97
The mean saturated fatty acid content of the samples (Table 14) was 45.1 mg/100 mg.
According to nutritional recommendations by health authorities, the content of saturated fatty
acids in margarines and table spreads should not exceed 30% in dietary fats (Brat et al.,
2000). However, in 16 samples the saturated fatty acids were higher. Obviously, partially
hydrogenated oils were replaced by palm oil or by oils from palm seeds (high in saturated
fatty acids) in some of these margarines. To increase the melting point of the margarines,
without hydrogenation, the manufacturers must increase the saturated fat content and/or
decrease the unsaturated fatty acid content. This is illustrated in Table 14 when comparing,
for example, Sample B and C (high saturated FA, low polyunsaturated FA) with a typical soft
margarine, Sample M (low saturated FA, high polyunsaturated FA). To keep to the
classification of a high unsaturated fat content and a higher melting point, the manufacturers
tend to keep the cis-9, 18:1 fatty acids as high as possible. This is well demonstrated in
Samples B, C, F, J and I (high mono-unsaturated FA, low polyunsaturated FA).
Table 14. The sum of the saturated, mono-unsaturated and polyunsaturated fatty acids (mg/ 100 mg) of the margarine samples analysed with a 120 m BPX-70 capillary column at a column temperature of 181o -1 C and a hydrogen gas flow rate of 30 cm sec
Sample Sat FA Mono FA Poly FA Trans FA TotalA 35.6 34.1 30.2 0.1 100B 52.2 38.5 9.3 0.1 100C 57.4 34.5 8.1 0.1 100D 27.6 31.3 22.6 18.4 100E 45.2 34.5 20.3 0.1 100F 53.2 37.6 9.2 0.0 100G 51.6 31.1 17.3 0.1 100H 48.8 32.7 18.5 0.1 100I 47.9 39.7 11.4 1.1 100J 47.6 38.3 11.8 2.3 100K 53.0 34.6 12.3 0.1 100L 48.2 34.4 17.4 0.0 100M 23.3 22.9 53.8 0.0 100N 46.0 41.0 13.0 0.0 100O 44.1 33.2 22.7 0.0 100P 34.7 34.7 30.6 0.0 100Q 49.9 34.5 15.6 0.0 100R 44.6 39.8 15.5 0.0 100
AVR 45.1 34.9 18.9 1.2STD 8.9 4.1 10.6 4.2
98
Table 14 shows that Sample D contains 72.4% unsaturated fatty acids and 27.6% saturated
fatty acids, giving it a much better unsaturated to saturated fatty acid ratio than for example,
Sample P, which contains 7.1% less unsaturated fatty acids and 7.1% more saturated fatty
acids. However, Sample D contains 18.4% trans unsaturated fatty acids, while Sample P
contains no trans fatty acids, making Sample D less desirable. During routine analyses using a
normal 30 m BPX-70 column, these trans isomers would not have been separated from the
cis-9 18:1 fatty acid and they would have been added to the mono-unsaturated fatty acids.
This would have given Sample D a high mono-unsaturated fatty acid composition of 49.8%
compared to only 34.7% for Sample P, but the real cis mono-unsaturated fatty acid
composition of Sample D is 31.3%, 1.4% less than Sample P. This illustrates that by using a
short BPX-70 column, the fatty acid composition of the samples can be misinterpreted.
The part of a chromatogram in Figure 23.1 shows the fatty acids’ composition of margarine
Sample P. All the main fatty acid peaks are well resolved. There are no detectable peaks
between 18:0 (peak A) and 18:1 (peak B), nor between 18:1 and 18:2 (peak C). The exception
is a small peak under the bracket, that was identified as cis-11, 18:1. This isomer is not caused
by partial hydrogenation, because it was also detected in blood samples that were analysed in
our laboratory using a 30 m BPX-70 column. These results illustrate that margarine, Sample
P, has no trans mono-unsaturated fatty acids.
181
oC
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500 A B C
c,11
181oC
minutes
milli
Vol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500 A B C
c,11
Figure 23.1. Part of the GLC chromatogram of sample P showing the 18:0 (A), 18:1 (B)
and 18:2 (C) fatty acids, as well as a cis-11 isomer (under the bracket)
99
The next part of a chromatogram (Figure 23.2) shows the fatty acids’ composition of
margarine, Sample D. The normal occurring fatty acids, peak A (18:0), peak B (18:1) and
peak C (18:2) are well resolved. This chromatogram clearly illustrates that this sample
contains unnatural cis and trans isomers (under the bracket) indicating that this margarine was
made from partially hydrogenated oil. Although not all the cis and trans isomers could be
baseline separated, the main isomers formed during partial hydrogenation can be identified
and quantified.
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
181 oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
181 oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
181 oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
181 oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
o
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
oC
A B C
0
500
1000
1500
2000
2500
A B C
minutes
mill
iVol
ts
25.00 27.50 30.00 32.50
0
500
1000
1500
2000
2500
A B C
Figure 23.2. Part of the GLC chromatogram of sample D showing the 18:0 (A), 18:1(B)
and 18:2 (C) fatty acids, as well as the cis and trans 18:1 fatty acid isomers (under the
bracket)
The results of the eighteen margarines and spreads illustrated that the selected group of South
African margarines had a lower trans fatty acid content than the margarines from some other
countries. For the evaluation of the total trans fatty acid consumption of the South African
population, the trans fatty acid content of the other staple foods also need to be determined.
100
CHAPTER 5
MATERIALS AND METHODS FOR CE
5.1 Introduction Fatty acids are usually determined by GLC. The methods incorporate derivatisation to obtain
volatility and detectability. The analytical time is usually long, because a very long capillary
column is needed for the identification of cis and trans fatty acids. Alternative methods
providing quicker analysis, preferably without a transmethylation step, are sought. A good
alternative method is CE that generally provides high efficiency and fast analyses.
Recently CE has been contemplated for the analyses of fatty acids in oils and fats, and several
approaches have been implemented towards such purpose (de Oliveira et al., 2003).
Separation is generally conducted in the zone electrophoresis mode, under indirect UV
detection, using a chromophore. It is also common practice to use large amounts of organic
solvents in the buffer, to enhance fat solubility. MEKC is an alternative mechanism for the
separation of fatty acids. For the separation of the cis and trans fatty acid isomers in
margarines, it was decided to use MEKC with direct UV detection and sodium dodecyl sulfate
(SDS) as the surfactant (Bohlin et al., 2003).
5.2 Samples
For the initial evaluation, a mixture of standard cis and trans FAMEs were used. (Nu- Chek-
Prep, INC. Elysian, Minnesota, USA). The standard was diluted in 99.5% ethanol and running
buffer was added. Finally, the mixture was degassed and mixed in a sonication bath.
101
5.3 Reagents and solutions
Sodium dodecyl sulfate, acetonitrile and boric acid were obtained from Aldrich (Sigma-
Aldrich (Pty) Ltd, PO Box 10434, Aston Manor 1630, South Africa) and urea from Sigma
(Sigma Chemical CO. St. Louis, MO 63178 USA). All the reagents were of analytical grade.
Stock solutions of electrolytic components were prepared by dissolving appropriate amounts
in Milli-Q water. Boric acid was adjusted to pH 9.2 with 20 M NaOH. The running buffer was
prepared with 24 mM SDS, 20% acetonitrile, 40 mM boric acid and 4 M urea in milli-Q
water. Prior to use, all the solutions were filtered through a 0.45 µm polypropylene filter.
5.4 Instrumentation
The experiment was conducted on a Hewlett-Packard CE system (Waldbronn, Germany),
equipped with a diode array detector and a temperature controlled capillary cartridge.
Detection was done with direct UV. The wavelength was set to 268 nm (Bohlin et al., 2003).
An uncoated 58 cm fused silica capillary, with 50 cm effective length, and 50 µm internal
diameter was used. The capillary was conditioned for 20 minutes with 2 M NaOH, 20 minutes
with 0.1 M NaOH and 10 minutes with Milli-Q water. Between runs, the capillary was
flushed for 2 minutes with 0.1 M NaOH, 2 minutes with organic modifier, 3 minutes with
water and 3 minutes with running buffer. Samples were injected by applying a pressure of
50 mbar for 10 seconds. The temperature was set to 15oC and the analysis was done with an
applied voltage of + 30 kV.
102
CHAPTER 6
CE RESULTS AND DISCUSSION
The FAMEs analysed are very hydrophobic and consists of three cis and three trans 18:1 fatty
acid isomers that differ only in position and geometry of the double bonds. This makes the
separation difficult without the presence of a pseudostationary phase in the running buffer.
SDS, as the surfactant, was tested as the pseudostationary phase. To enhance the solubility of
the hydrophobic fatty acids, urea was added to the running buffer. Without urea, the fatty
acids will precipitate (Bohlin et al., 2003). No separation was achieved using 24 mM SDS
and the SDS concentration was increased to 35 mM. A small peak was detected, but no
separation. Since no baseline separation was achieved with different concentrations of SDS
buffer, a longer capillary was utilised. Theory predicts that a longer capillary leads to longer
migration times, and hence, band broadening because of longitudinal diffusion. The small
peak did indeed elute much later, but without separation. It was expected that at least two
peaks, a cis and a trans peak would be separated.
After the initial experiments, it was concluded that the separation of cis and trans 18:1 fatty
acid isomers will not be possible with only one surfactant. Therefore, the use of different
surfactants need to be investigated. The detection was done by direct UV detection, but
indirect UV detection with chromophores also needs to be investigated further. The standard
FAME mixture that was used comprises long chain fatty acids that are poorly soluble in
aqueous solutions and the use of different non-aqueous buffers need to be investigated. Erim
et al. (1995) reported that the longer chain fatty acids dissolve considerably better with a
buffer containing 75% acetonitrile. Under these conditions, fatty acids with chain lengths up
to 19 carbons could be completely separated.
Capillary electrochemistry (CEC) is an interesting alternative to the normal CE. In CEC, the
mobile phase flow is generated electrokinetically by using a high voltage. Instead of the fused
silica open tube capillary used in CE, a capillary packed with a stationary phase is used for
selectivity. CEC combines the efficiency and analytical speed of CE with the selectivity GLC.
103
The use of CE methodologies, for the identification and quantification of the different cis and
trans fatty acid isomers in partially hydrogenated oils, is a possibility, but plenty more
research is needed. As the standardisation and optimisation of this technique fall outside the
scope of this study, the comparison between the two methodologies could not be done.
104
CHAPTER 7
CONCLUSIONS
The first important step in the identification and quantification of the different cis and trans
fatty acid isomers in food samples, is the correct extraction of all the fatty acids. Of the three
different extraction solutions evaluated in this study, it can be concluded that the
chloroform/methanol (2:1) solution gave the best fatty acid recovery. Therefore, this solution
can be recommended for the extraction of the fatty acids in margarines.
Before the extracted fatty acids can be analysed by GLC, it is necessary to convert them to
low molecular weight non-polar derivatives, such as methyl esters. In this study, two
transmethylation reagents were evaluated and it was found that a solution of 5% concentrated
sulphuric acid in double distilled methanol gave a 98.5 ± 0.35% FAME recovery using a 120
m capillary column. The FAME recovery was about 7% lower when using a 0.5 M sodium
methoxide in methanol solution, and it was concluded that there was some sample loss or
incomplete transmethylation when using 0.5 M sodium methoxide in methanol. A recovery
loss was also observed when the ratio of sample concentration to volume of transmethylation
reagent increased. When working with samples of unknown fatty acid concentration it is
recommended that an acid-catalysed transmethylation solution be used.
The column length plays a critical role in the analyses of cis and trans isomers in the samples.
Although a 30 m BPX 70 column gave good separations with a short analytical time for
routine sample analyses, the cis and trans fatty acid isomers in the samples, especially those
occurring between 18:0 and 18:2, could not be separated. Even the use of different column
temperatures and column gas flow rates did not improve separation. The analytical elution
range of the different cis and trans isomers of 18:1 were contracted into a too narrow
retention time range for proper identification.
The 120 m BPX-70 column provided the required mechanism for extending the retention
times of the different isomers by retaining the different compounds longer. In this way, the
105
retention times of the different isomers were pulled apart, and a greater separation space was
available to the different isomers.
Different column temperatures have a major impact on the separation power of the column.
Isothermal operation at 181oC produced the least overlapping peaks, but some of the isomers
will always overlap using a BPX-70 column regardless of the column temperature you use.
Isothermal operations above or below 181oC produced some additional isomer overlapping
problems.
Except for a change in the retention times, there was very little effect on the resolution and
column efficiency with the different gas velocities when hydrogen was utilised. Therefore, the
precise flow rate was less critical. Although hydrogen is highly flammable, it remains the best
carrier gas to use for the analyses of fatty acids because of its high diffusivities and low
resistance to mass transfer. Even when using nitrogen or helium as a carrier gas, hydrogen is
still needed for the FID flame.
The labelling law specifies that only the total concentration of the trans fatty acids must be
displayed on the food label, though by using a long highly-polar capillary column, the
different cis and trans isomers could be resolved making it possible to identify and quantify
most of the isomers. From the concentrations of the different isomers, a very good prediction
can be made of the source of the oils and fats that were used in the preparation of the
margarines.
The Ag-TLC fractionation of the samples into the cis and trans 18:1 fatty acid isomers
remains a good method prior to GLC analyses, to identify the overlapping cis and trans
isomeric groups. However, this method did not separate all the different overlapping isomers.
Two negative aspects of this method are the difficulty to quantify the different isomers, and
the very laborious technique. With TLC, there is always a possibility of loosing some of the
fatty acids during the long and laborious analytical preparation steps.
The results of the different margarines analysed, were surprising. Of the 18 samples analysed,
only one had a high trans fatty acid concentration. Overall, local margarines do not have trans
fatty acid content as high as that reported for some other countries. Although the trans fatty
106
acid content of the selected group of margarines was low, the total saturated fatty acid content
was higher than the recommended percentage content. In order to calculate the total intake of
trans fatty acids more data is needed, particularly on the trans fatty acid content of fast foods
and other baked products made from shortenings. An advantage of using GLC fitted with a
long highly-polar capillary column to analyse foods prepared from partially hydrogenated oils
and fats, is that all the normal occurring and most of the cis and trans fatty acids in the
samples can be identified and quantified in one analytical run.
An accurate determination of trans fatty acid intakes of individuals will only be possible after
food composition tables have been updated with the trans fatty acid content of the different
foods consumed in our country. This information is essential to study the effect of trans fatty
acids in epidemiological studies, and to explore the potential negative effects specific trans
isomers in certain foods have on different diseases.
In conclusion, this study demonstrated that the major cis and trans isomers can be identified
and quantified by GLC using a 120 m BPX-70 capillary column. Although these capillary
columns improved the separation, not all overlapping isomers could be separated in one
isothermal column temperature run. To identify more of the overlapping isomers, experiments
with different types of columns should also be conducted.
107
CHAPTER 8
REFERENCES
Adlof, R.O., 1994, Separation of cis and trans fatty acid methyl esters by silver ion HPLC. J.
Chromatogr. A., 659, 95-99.
Almendingen K., Jordal O., Kierulf P., Sandstad B., Pedersen J., 1995, Effects of partially
hydrogenated fish oil, partially hydrogenated soybean oil and butter on serum lipoproteins and
Lp(a) in men. J. Lipid Res., 36, 1370-1384.
Department of Health and Human Services, 2003, American Food and Drug Administration:
Final rule on trans fatty acids and nutrition labeling. [On line]. Available:
http://www.cfsan.fda.gov/~lrd/fr03711a.html. [2006, November 10].
Anderson J.T., Grande F., Keys A., 1961, Hydrogenated fats in the diets and lipids in the
serum of man. J. Nutr., 75, 388-394.
Aro A., Kosmeijer-Schuil T., Van De Bovenkamp P., Hulshof P., Zock P., Katan M.B., 1998,
Analyses of C18:1 cis and trans fatty acid isomers by the combination of Gas-liquid
Chromatography of 4,4 –Dimethyloxazoline derivatives and methyl esters. J. Am. Oil Chem.
Soc., 75, 977- 985.
Ascherio A., Willet W.C., 1997, Health effect of trans fatty acids. Am. J. Clin. Nutr., 66,
1006-1010.
Ball M.J., Hackett D., Duncan A., 1993, Trans fatty content of margarines, oils and blended
spreads available in New Zealand. Asia Pac. J. Clin. Nutr., 2, 165-169.
Belluzzi A., Brignola C., Campeiri M., Pera A., Boschi S., Miglioli M., 1996, Effect of an
enteric-coated fish oil preparation on relapses in Crohn’s disease. N. Engl. J. Med., 334, 1557-
1560.
Belury M.A., 2002, Not all trans fatty acids are alike: what consumers may lose when we
oversimplify nutritional facts. J. Am. Diet. Assoc., 102 (11), 1606-1607.
108
Berdeaux O., Juaneda P., Sebedio J.L., 1998, Analysis of conjugated and trans fatty acids
after derivatization. Analusis, 26, M45-M51.
Bligh E.G., Dyer W.J., 1959, A rapid method for total lipid extraction and purification. Can.
J. Biochem. Physiol., 37, 911-917.
Bohlin M.E., Ohman M., Hamberg M. Blomberg L.G., 2003, Separation of conjugated
trienoic fatty acid isomers by capillary electrophoresis. J. Chromatogr. A., 985, 471-478.
Brat J., Pokorny J., 2000, Fatty acid composition of margarines and cooking fats available on
the Czech market. J. Food Comp. Analyses, 13, 337-343.
Cabrini L., Landi L., Stefanelli C., Barzanti V., Sechi A.M., 1992, Extraction of lipids and
lipophilic antioxidants from fish tissues: A comparison among different methods. Comp.
Biochem. Physiol., 101B, 383-386.
Castaneda G., Rodriguez-Flores J., Rios A., 2005 Jun, Analytical approaches to expanding the
use of capillary electrophoresis in routine food analyses. J. Sep. Sci., 28(9-10), 915-24.
Christie W.W In., Sebedio J.L., Christie W.W., Adolf R., editors., 2003, Advances in
conjugated linoleic acid research. Vol. 2, AOAC Press, Champaign (IL), 1-36.
Christie WW., 1989, Gas chromatography and lipids. A practical guide. The oily press. AYR,
Scotland.
Christie W.W., 1990, Preparation of methyl esters-Part 1. Lipid Technology, 2, 48-49.
Christie W.W., 1994, Why I dislike boron trifluoride- methanol. Lipid Technology, 6, 66-68.
Christie W.W., (ed.) 1993, Advances in Lipid Methodology: Preparation of lipid extracts from
tissues. Oily Press: Dundee, 195-213.
Christie W.W., 1972, Methylation of fatty acids. In: F.D. Gunstone, (ed).Topics in Lipid
Chemistry. Vol. 3. Logos Press: London, 171-197.
Christopherson S.W., Glass R.L., 1969, Preparation of milk fat methyl esters by alcoholysis in
an essentially nonalcoholic solution. J. Dairy Sci., 52, 1289-1290.
109
De Koning S., Van der Meer B., Alkema G., Janssen H., Brinkman U., 2001, Automated
determination of fatty acid methyl esters and cis/trans methyl ester composition of fats and
oils. J. Chromatogr. A., 922, 391-397.
De Oliveira M.A.L., Solis V.E.S., Gioielli L.A., Polakiewicz B., Tavares M.F.M., 2003,
Method development for the analyses of trans-fatty acids in hydrogenated oils by capillary
electrophoresis. Electrophoresis, 24, 1641-1647.
Deman L., Deman J.M., 1983, Trans fatty acids in milk fat J. Am.Oil Chem.Soc., 60, 1095-
1098.
Duchateua G.S.M.J.E., Van Oosten H.J., Vasconcellos M.A., 1996, Analyses of cis and trans
fatty acid isomers in hydrogenated and refined vegetable oils by capillary Gas Liquid
chromatography. J. Am. Oil Chem. Soc., 60, 1788-1793.
Dutton H.J., 1997, Hydrogenation of fats and its significance. In: Eraken E.A., Dutten HJ.,
Geometrical and positional fatty acid isomers. Champaign, IL: American Oil Chemists
Society, 1-16.
Emkin E.A., 1995, Trans fatty acids and coronary heart disease risk: physiochemical
properties, intake and metabolism. Am. J. Clin. Nutr., 62, 659s-668s.
Emkin E.A., 1984, Nutrition and biochemistry of trans fatty acid isomers in hydrogenated
oils. Annu. Rev. Nutr., 4, 339-376.
Emkin E.A., 1979, Utilization and effects of isomeric fatty acids in humans. In; Emkin E.A.,
Dutton H.J., eds. Geometrical and positional fatty acid isomers. Champaign, IL: American Oil
Chemists Society., 99-129.
Erim F.B., Xu X., Kraak J.C., 1995, Application of micellar electrokinetic chromatography
and indirect UV detection for the analysis of fatty acids. J. Chromatogr. A., 694, 471-479.
Firestone D., Labouliere P., 1965, Determination of isolated trans isomers by infrared
spectroscopy. J. Assoc. Anal. Chem., 48, 437-443.
Folch J., Lees M., Sloane S.G.H., 1957, A simple method for the isolation and purification of
total lipids from animal tissue. J. Biol. Chem., 226, 497-509.
110
Fulk W.K., Shorb MS., 1970, Production of an artifact during methanolysis of lipids by boron
trifluoride-methanol. J. Lipid Res., 11, 276-277.
General Conference Nutrition Council, 2002, Position Statements: Trans fatty acids: how safe
are they? [On line]. Available: http://www.Andrews.edu/NUFS/trans.html. [2006, November
13].
Glew R.H., Herbein J.H., Ma I., Obadofin M., Wark M.A., Van der Jagt D.J., 2006, The trans
fatty acid and conjugated linoleic acid content of Fulani butter oil in Nigeria. J. Food Comp.
Analyses, 19, 704-710.
Hay J.D., Morrison W.R., 1970, Isomeric monoenoic fatty acids in bovine milk fat. Biochim.
Biophys. Acta., 202, 237-243.
Heiger D., 1992, High performance capillary electrophoresis- An Introduction. 2nd Edition.
Printed in France. Hewlett-Packard Company.
Innis S.M., Green T.J., Halsey T.K., 1999, Variability in the trans fatty acid content of foods
within a food category: Implications for estimation of dietary trans fatty acid intakes. J. Am.
Coll. Nutr., 18, 255-260.
Ip C., Scimeca J.A., Thompson H.J., 1994, Conjugated linoleic acid. A powerful
anticarcinogen from animal fat source. Cancer, 74, 1050-1054.
Jamieson G.R., Reid E.H., 1969, The analyses of oils and fats by Gas chromatography II, the
alkaline isomerisation of linoleic acid. J. Chromatogr. A., 20, 232- 239.
Juanèda P., 2002, Utilization of reverse-phase high-performance liquid chromatography as an
alternative to silver-ion chromatography for the separation of cis and trans C18:1 fatty acid
isomers. J. Chromatogr. A., 954, 285-289.
Kang J.X., Wang J.A., 2005, simplified method for analyses of polyunsaturated fatty acids.
BMC Biochemistry, 6, 5-11.
Katan M.B., Mensink R.P., Zoak P.L., 1995, Trans fatty acids and their effects on lipoproteins
in humans. Annu. Rev. Nutr., 15, 473-493.
111
Kepler C., Hirons K., Mc Neill J., Tove S., 1966, Intermediates and products of the
biohydrogenation of linoleic acid by Butyrivibrio fibrisolvents. J. Biol. Chem., 241, 1350-
1354.
Keweloh H., Heipieper H.J., 1996, Trans unsaturated fatty acids in bacteria. Lipids, 31, 129-
137.
Kohlmeier L., Simonsen N., van’t Veer P., Strain J.J., Martin-Moreno J.M., Margolin B.,
Hutttunen J.K., Fernandes-Crehuet Navajas J., Martin B.C., Thamm M., Kardinaal A.F., Kok
F.J., 1997, Adipose tissue trans fatty acids and breast cancer in the European Community
Multicentre study on antioxidants, Myocardial Infarction and Breast cancer. Cancer
Epidemiol. Biomark. Prev., 6, 705-710.
Koletzko B., 1992, Trans fatty acids may impair biosyntheses of long chain polyunsaturates
and growth in man. Acta Paediatr., 81, 302-306.
Kramer J. K.G., Cruz-Hernandez C., Deng Z., Zhou J., Jahreis G., Dugan M.E.R., 2004,
Analysis of conjugated linoleic acid and trans 18:1 isomers in synthetic and animal
products1,2,3,4 Am. J. Clin. Nutr., 79(6), 1137S-1145S.
Lake R., Thomson B., Devane G., Scholes P., 1996, Trans fatty acid content of selected New
Zealand. Foods. J. Food Comp. Analyses, 9, 365-347.
Lamberto M., Ackman R., 1994, Confirmation by gas chromatography/mass spectrometry of
two unusual trans-3-monoethylenic fatty acids from the Nova Scotia seaweeds Palmaria
palmata and Chororidrus crispus. Lipids, 29, 441-444.
Ledoux M., Laloux L., Wolff R.L., 2000, Analytical methods for determination of trans –C18
fatty acid isomers in milk fat. A review. Analusis, 28, 402-412.
Lepage G., Claude C.R., 1984, Improved recovery of fatty acids through direct
transesterification without prior extraction and purification. J. Lipid Res., 25, 1391-1396.
Lichtenstein A.H., Ausman L.M., Jalbert S.M., Schaefer E.J., 1999, Effects of different forms
of dietary hydrogenated fats on serum lipoprotein cholesterol levels. N. Engl. J. Med., 329,
1969-1970.
112
Lippi G., Guidi G., 1999, Biochemical risk factors and patient’s outcome: The case of
Lipoprotein (a). Clin. Chim. Acta., 280, 59-71.
Lough A.K., 1964, The production of methoxy-substituted fatty acids as artifacts during the
etherification of unsaturated fatty acids with methanol containing boron trifluoride. Biochem.
J. 90: 4c-5c.
Mackie R, White B, Bryant M. 1991. Lipid metabolism in anaerobic systems. CRC. Crit. Rev.
Microbiol., 17, 449-478.
Mahfouz M.M., Smith T., Kummerrow F.A., 1984, Effect of dietary fats on desaturase
activities and the biosynthesis of fatty acids in rat liver microsomes. Lipids, 19, 214-222.
Marinetti, G.V., 1966, Low temperature partial alcoholysis of triglycerides. J. Lipid Res., 7,
786-788.
McMurry John, 2000, Organic Chemistry. Brooks/Cole: London.
Mensink R.P., Kata M.B., 1990, Effect of dietary trans fatty acids on high–density and low-
density lipoprotein cholesterol levels in healthy subjects. N. Engl. J. Med., 323, 439-445.
Mensink R.P., Kata M.B., 1992, Effect of dietary fatty acids on serum lipids and lipoproteins:
meta-analyses of 27 trails. Arterioscler. Thromb., 12, 911-919.
Molkentin J., Precht D., 1995, Optimized analyses of trans octadecenoic acids in edible fats.
Chromatographia, 41, 267-272.
Morrison W.R., Smith L.M., 1964, Preparation of fatty acid methyl acids and dimethylacetals
from lipids with boron fluoride-methanol. J. Lipid Res., 5, 600-608.
Nicolosi R.J., Rogers E.J., Kritchevsky D., Scimeca J.A., Huth P.J., 1997, Dietary conjugated
linoleic acid reduces plasma lipoproteins and early arteriosclerosis in hypercholesterolemic
hamsters. Artery, 22, 266-277.
NutritionData, 2006, Fatty Acids [On line]. Available: http://www.nutritiondata.com/fatty-
acids.html. [2006, November 13].
Oils and Fats, 2005, Malaysian Palm Oil Promotion Council. Vol. 2, Issue 3.
113
Oslund-Lingvist A., Albanus L., Croon L., 1985, Effect of dietary trans-fatty acids on
microsomal enzymes and membranes. Lipids, 20, 620-624.
Parodi P.W., 1997, Cows’ milk fat components as potential anticarcinogenic agents. J. Nutr.,
127, 1055-1060.
Parodi P.W., 1976, Composition and structure of some consumer available edible fats. Ibid,
53, 530-534.
Parodi P.W., 1976, Distribution of isomeric octadecenoic fatty acids in milk fat. J. Dairy Sci.,
50, 1870-1873.
Paterson H.B., 1996, Hydrogenation of fats and oils. AOCS Press: Champaign, Illinois.
Perkin E.G., 1991, Analyses of fats, oils and lipoproteins. AOCS Press: Champaign, Illinois.
Precht D., Molkentin J., 1996, Rapid analysis of the isomers of trans octadecenoic acid in
milk fat. Int. Dairy J., 6, 791-809.
Precht D., Molkentin J., 1997, Effect of feeding on conjugated cis D9, trans D11-
octadecadienoic acid and other isomers of linoleic acid in bovine milk fats. Nahrung, 41, 330-
335.
Radin N.S., 1981, Extraction of tissue lipids with a solvent of low toxicity. Methods
Enzymol., 72, 5-7.
Ratnayake W.M.N., Hansen S.L., Kennedy M.P., 2006, Evaluation of the CP-Sil 88 and SP-
2500 GC columns used in the recently approved AOAC official method Ce 1h-05:
Determination of cis-, trans-, saturated, monounsaturated and polyunsaturated fatty acids in
vegetable or non-ruminant animal oils and fats by capillary GLC method. J. Am. Oil Chem.
Soc., 83(6), 475-488.
Ratnayake W.M.N., Plouffe L.J., Pasquier E., Gagnon C., 2002, Temperature sensitive
resolution of cis and trans fatty acid isomers of partially hydrogenated vegetable oils on SP-
2560 and CP-Sil 88 capillary columns. J. Am. Oil Chem. Soc., 85(5), 1112-1118.
114
RatnayakeW.M.N., 2004, Overview of methods for the determination of trans fatty acids by
gas chromatography, silver-ion thin-layer chromatography, silver-ion liquid chromatography,
and gas chromatography/mass spectrometry. J. Ass. Off. Anal. Chem. Int., 87, 523-539.
Richardson R.K., Fong B.Y., Rowan A.M., 1997, The trans fatty acid content of fats in some
manufactured foods commonly available in New Zealand. Asia Pac. J. Clin. Nutr., 6(4), 239-
245.
Sanders T.A., Oakley F.R., Miller G.J., Mitropolis K.A., Crook D., Oliver M.F., 1997,
Influence of n-6 versus n-3 polyunsaturated fatty acids in diets low in saturated fatty acids on
lipoproteins and haemostatic factors. Arterioscler. Thromb. Vasc. Biol., 17, 3449-3460.
Scimeca J.A., Miller G.D., 2000, Potential health benefits of conjugated linoleic acid. J. Am.
Coll. Nutr., 9(4), 470s-471s.
Semma M., 2002, Trans fatty acids: Properties, Benefits and Risks. J. Health Sci., 48, 7- 13.
Sommerfeld M., 1983, Trans unsaturated fatty acids in natural products and processed foods.
Prog. Lipid Res., 22, 221- 233.
South Africa, 2002, Draft Regulations Relating to Labelling and Advertising of Foodstuffs,
no. R 1055/2002. Pretoria: Govt. Printers.
Stoffel W., Wu F., Ahrens E.H., 1959, Analyses of long chain fatty acids by Gas-Liquid
Chromatography. Anal. Chem., 31, 307-308.
St-Onge M.P., Jones P.J.H., 2002, Physiological effect of medium-chain triglycerides:
Potential agents in the prevention of obesity. J. Nutr., 132, 329-332.
Strampfer M., 2004, Primary Prevention of heart disease through lifestyle and diet. Public
lecture series. Report. Linus Pauling Institute.
Sundram K., Ismail A., Hayes K.C.M., Jeyamalar R., Pathmanathan R., 1997, Trans fatty
acids adversely affect the lipoprotein profile relative to specific saturated fatty acids in
humans. J. Nutr., 127, 514S-520S.
Tang T.S., 2002, Fatty acid composition of edible oils in the Malaysian market, with special
reference to trans- fatty acids. J. Palm Oil Res., 14(1), 1-8.
115
Technical Committee of the Institute of Shortening and Edible Oils, Inc., 2006, Food Fats
and Oils. Ninth Edition. Institute of Shortening and Edible Oils. Washington, DC 20006. [On
line]. Available: http://www.iseo.org/FoodFatsOils2006.pdf. [2006, November 10].
Thompson R.H., 1997, Direct measurement of total trans- and cis-octadecenoic fatty acids
based on a gas-liquid chromatographic class separation of trans-18:1 and cis-18:1 fatty acid
methyl esters. J. Chromatogr. Sci., 35, 536-544.
Ulbrecht F., Henninger M., 1996, estimation of trans fatty acidic content of edible oils and
fats: an overview of analytical methods. Eur. J. Med. Res., 1, 94-99.
Ulbrecht F., Henninger M., 1994, Quantitation of trans-fatty acids in milk fat using
spectroscopic and chromatographic methods. J. Dairy Res., 61, 517-527.
Valcarcel M., Rios A., Arce L., 1989, Coupling Continuous Sample Treatment Systems to
Capillary Electophoresis. Crit., Rev. in Anal.Chem. 28, 63-81.
Vidgren H.M., Louheranta A.M., Agren J.J., Schwab U.S., Uusitupa M.I.J., 1998, Divergent
incorporation of dietary trans fatty acids in different serum lipid fractions. Lipids, 33, 855-
962.
Willet J.E., 1987, Gas Chromatography. John Wiley & Sons: Inc, Somerset, New Jersey,
U.S.A.
Willet W.C., Stampfer M.J., Manson J.E., Colditz G.A., Speizer F.E., Sampson L.A.,
Henneken C.H., 1995, Intake of trans fatty acids and risk of coronary heart disease among
women. Lancet, 341, 581-585.
Wolff R.L., 1995, Content and distribution of trans 18:1 acids in ruminant milk and meat fats.
Their importance in European diets and their effect on human milk. J. Am. Oil Chem. Soc.,
72, 259-272.
Wolff R.L., Bayard C.C., 1995, Improvement in the resolution of individual trans-18:1
isomers by capillary gas-liquid chromatography - use of a 100m CP-Sil 88 column. J. Am. Oil
Chem. Soc., 72, 1197-1201.
116
top related